sign in sign up
<<
Home , 12 BC, Acholla, Apollonia (Sicily), Blanda (city), Bruni (horse), Caulonia (ancient city)

UC Berkeley UC Berkeley Electronic Theses and Dissertations

Title The Monumental Villa at Palazzi di and the Roman Elite in () during the Fourth Century AD

Permalink https://escholarship.org/uc/item/2jf6g36f

Author Bruni, Maria Gabriella

Publication Date 2009

Peer reviewed|Thesis/dissertation

eScholarship.org Powered by the California Digital Library University of California

The Monumental Villa at Palazzi di Casignana and the Roman Elite in Calabria (Italy) during the Fourth Century AD.

by

Maria Gabriella Bruni

A dissertation submitted in partial satisfaction of the

Requirements for the degree of

Doctor of Philosophy

in

Classical Archaeology

in the

GRADUATE DIVISION

of the

UNIVERSITY OF CALIFORNIA

Committee in Charge

Professor Christopher H. Hallett, Chair Professor Ronald S. Stroud Professor Anthony W. Bulloch Professor Carlos F. Noreña

Fall 2009

The Monumental Villa at Palazzi di Casignana and the Roman Elite in Calabria (Italy) during the Fourth Century AD.

Copyright 2009

Maria Gabriella Bruni Dedication

To my parents, Ken and my children.

i

AKNOWLEDGMENTS

I am extremely grateful to my advisor Professor Christopher H. Hallett and to the other members of my dissertation committee. Their excellent guidance and encouragement during the major developments of this dissertation, and the whole course of my graduate studies, were crucial and precious. I am also thankful to the Superintendence of the Archaeological Treasures of for granting me access to the site of the Villa at Palazzi di Casignana and its archaeological archives. A heartfelt thank you to the Superintendent of Claudio Sabbione and to Eleonora Grillo who have introduced me to the villa and guided me through its marvelous structures. Lastly, I would like to express my deepest gratitude to my husband Ken, my sister Sonia, Michael Maldonado, my children, my family and friends. Their love and support were essential during my graduate studies at UC Berkeley.

ii

Table of Content

Introduction 1

Chapter one: The geography and a brief historical survey of Bruttii (Calabria, Italy) 4

Chapter two: The layout of the Villa at Palazzi di Casignana 10

Chapter three: The archaeological area of the villa 19

Chapter four: The monumentality phase of the villa in the fourth century AD 30

Chapter five: The 87

Chapter six: The late imperial villas near Locri 142

Chapter seven: Rural estate and urban context 161

Chapter eight: The economy of the fourth century elite and that of Bruttii 171

Chapter nine: Room I and its function in the fourth century AD 205

Chapter ten: Christianity in fourth century Bruttii 263

Conclusions 282

Bibliography 285

iii

Abstract

The Monumental Villa at Palazzi di Casignana and the Roman Elite in Calabria (Italy) during the Fourth Century AD. by

Maria Gabriella Bruni

Doctor of Philosophy in Classical Archaeology

University of California, Berkeley

Professor Christopher H. Hallett, Chair

In the fourth century AD, the early imperial Villa at Palazzi di Casignana, located on the Ionian coast, 15 km south of the colony of Locri (Italy) was rebuilt on a grand scale. The addition of two sets of baths, apsidal rooms, lavish wall marble veneers, and exquisite floor mosaics made it comparable to other important Late Antique villas, such as that at Piazza Armerina in . This study examines the architecture and decorative apparatus of the villa in order to better understand the historical and cultural context of this fourth century remodeling. Comparative analysis with other Roman villas from the same period leads to some obvious and provocative hypotheses with regard to the function of the rooms, and the role of the surrounding estate. The opulence of the villa and the 15 hectares of land that surrounds it, are evidence that Locri and its territory hosted a wealthy elite in the fourth century AD. This study argues against a long-held belief in the field of archaeology and history that Bruttii was impoverished or insignificant during the late imperial period. Archaeological data from the region, in particular from the Villa at Palazzi di Casignana, has provided an incentive for us to re-examine ancient and modern sources in order to reassess the economic fortunes of Bruttii at this time. It is likely that the abundance of natural resources of the region prompted the late Roman elite to invest their fortunes in this part of the Empire. Literary and archaeological evidence suggests that wine-making was among the most profitable sources of revenue for the late Roman elite in Bruttii. A large number of wine amphorae (Keay 52) have been unearthed in the region, strongly suggesting that there was a large-scale production of wine in this area. Recent studies contend that these amphorae were manufactured in Bruttii, and there is ample archaeological evidence showing that they were exported to distant regions of the Empire, including and France. Even at the Villa of Palazzi di Casignana, a large number of these amphorae have been discovered, indicating that the villa may have been an important local hub for wine-making. In addition, the largest and most lavish room of the Villa at Palazzi di Casignana, Room I, which is inserted in the seafront sector of the villa, is considered at length. By comparing this Room to similar ones found in other late Roman villas, this study investigates the possible functions of this space and the likely role that the villa had in the area of Locri during the fourth century AD. Although there have been extensive excavations at the villa’s site during the past few decades, and a detailed archaeological report has been recently published, it is premature to draw definite conclusions about the function and ownership of this outstanding complex. There is no doubt that the Villa at Palazzi di Casignana stands as a significant evidence that in Locri and in its territory a wealthy elite resided during the later centuries of the Roman history and that Roman Calabria is worthy of deeper and broader exploration.

1

INTRODUCTION

An enormous imperial Roman villa scenically located on the Ionian coast of Bruttii (modern Calabria), 15 km south of the ancient Greek colony of Locri, was discovered at Palazzi di Casignana in 1964. Archaeological evidence from the site combined with new scholarly work illuminates the role played by the region during fourth century AD. In addition, this study disproves the long-held belief that this southern region of the Empire was impoverished and insignificant during the late Roman times. The Villa at Palazzi di Casignana spans 15 hectares and is exceptional in for its size, lavish floor mosaics and walls covered with marble veneers. More than 20 rooms are embellished with a superb variety of geometric and figural designs, representing the largest surviving collection of Roman floor mosaics in Bruttii (Calabria). The villa stands 300 meters from the and is the nucleus of an extensive archaeological area worthy of much more scientific exploration. It was originally built in the first century AD, but rose towards a greater monumentality and lavishness between the end of the third and the first half of the fourth century AD, enduring through multiple significant renovations. In the fifth century, the villa ceased to be occupied for unknown reasons. The opulence displayed at the Villa at Palazzi di Casignana in the fourth century AD is in stark contrast with the traditional view that Bruttii was insignificant and underwent an irreparable decline in Roman times, particularly during the last centuries of the Empire. This study focuses on the fourth century AD phase of the villa and attempts to answer the most important questions raised by this complex. Who owned this complex and what was its function in the fourth century AD? What were the sources of wealth in Bruttii that may have prompted the enhancement and monumentality of the villa during this time? How involved was Bruttii in the development of the late Roman economy during the last centuries of the ? What can we infer from the villa’s architectural and decorative elements to provide a better understanding of how the Roman elite lived and operated in Bruttii in the fourth century AD? In order to provide answers to these and related questions, this study is organized in two parts. The first part looks closely at the architecture and decorative apparatus of the villa (Chapter 1-5). The second part focuses on the society and economy of late imperial Bruttii (Chapter 6-10). Chapter 1 explains the geography and provides a brief historical survey of Calabria. Chapter 2 analyzes the villa’s layout and the property associated with it. Chapter 3 illustrates the archaeological area around the villa where buildings and burials were uncovered to show that the villa was the center of a rural settlement. In this section evidence is provided to argue against the hypothesis advanced by archaeologists that the villa was a statio. Chapter 4 describes the villa’s renovations during the fourth century AD, when the building reached its greatest level of monumentality and lavishness, and provides an interpretation of each room’s function in proper historical and social context. Chapter 5 analyzes the floor mosaics that decorate the villa, which represent the largest collection of Roman mosaics in the region. The majority of the mosaics inside the villa are geometric and a handful is figural. Mosaics are an invaluable tool for retrieving the function of interior spaces inside Roman dwellings. This study examines how the owner of Villa at Palazzi di Casignana favored particular figural or ornamental designs when decorating his residence, and used these designs to decorate specific rooms according to the function of each room. Figural scenes and geometric patterns of floor mosaics offer a wide range of information about the visual culture, and the economic and social identity of people who commissioned mosaics. The second part of the dissertation starts with Chapter 6, which surveys

1

other late Roman villas discovered in the territory of Locri. These villas point to the vitality of the Roman upper classes in the region and they shed a light on the economic circumstances in Bruttii during the later centuries of the Empire. Chapter 7 examines the relationship between the country and the town during the late imperial period, particularly in the area of Locri. Chapter 8 explores a wide range of literary texts to help reconstruct the economic and social circumstances during the fourth century AD in Bruttii, especially in the territory of Locri, to understand what prompted dramatic transformations and enhancements of the early imperial villa at Palazzi di Casignana. Chapter 9 describes Room I, the largest room of the villa with an apsidal cruciform plan, and considers its role during the fourth century AD. Alhough Room I may have been used as a dining room, or an audience hall, this study is more inclined to believe that it was a private or semi-private chapel for hosting religious activitites. Literary and archaeological evidence of private Christian worship at this time throughout the Empire suggests that Room I may have hosted a religious activities in the fourth century AD. The diffusion of Christianity in Bruttii and in the area of Locri during this time supports this hipothesis, and this is the subject of Chapter 10. Archaeologists have suggested that Villa at Palazzi di Casignana might have been the nucleus of a statio, namely a travel post of the cursus publicus. This study is more inclined to believe that the villa was a private dwelling of a very wealthy owner in the fourth century AD because of the lavish architectural features and the seigniorial character of this complex. Further excavations on the site could provide much needed insight to better explain the villa’s function during this period, including any relation it might have had with the diffusion of Christianity. The enormous size and monumentality of the villa are features not necessarily associated with a statio. Late Roman sources, such as Palladius and archaeological data corroborate architectural trends across the Empire during the fourth century AD. Roman villas expanded reaching the size of towns in the fourth and fifth centuries. By combining otium (leisure) with negotium (business) even more than in earlier times, these villas were retreat residences involved in farming and trade and became self-sufficient economic centers. Buildings near the Villa at Palazzi di Casignana perhaps included dwellings of peasants, facilities for the production and storage of agricultural goods such as wine, olive oil, fruit, or stables for cattle rearing attesting that the villa’s owner was likely involved in farming and trade. The Villa at Palazzi di Casignana is clearly a manifestation of immense wealth. The economy of Bruttii is successful at this time. The availability of natural resources in the region encouraged the Roman elite to invest a fortune in this region. The villas near Palazzi di Casignana, in the territory of Locri, stand as a proof that Bruttii during the late imperial time was led by a lively and active elite who exploited the natural resources of the region to increase their wealth, being favored also by an articulated road system and access to important maritime ways through major Roman harbors. Even the monetary economy, based on the gold solidus, introduced in the early fourth century by Constantine, greatly contributed to the late classical prosperity of the Roman elite in Bruttii (as it did in other regions of the Roman world). The historical and economic circumstances of this period might have driven Bruttii to become a significant economic region rivaling other successful regions in the Empire. At the time when Villa at Palazzi di Casignana expanded in size and sumptuousness several villas across the Empire experienced similar transformations: the remodeling of various rooms, the addition of at least a bath-suite, one or more reception halls, large apsidal rooms, floors, expensive decorations and statuary. These changes signify more than average renovations or redecoration. The late Roman villas at Piazza Armerina (Sicily), Desenzano (Northern Italy), Faragola (Puglia) and those outside of Italy in Spain, Southwest France, and the Balkans, all seem

2 to play even a greater social role than before. These residences were “the forum made private” rather than retreats from public life (Brown 1992, p.273). They denote the elite’s desire to exhibit wealth in an ostentatious manner and are telling about the role that the villa played at this time in the elite’s life and culture. It is the intent of this research, through a detailed analysis of Villa at Palazzi di Casignana’s architecture and decoration, to bring a better understanding of the role that the Roman elite of Bruttii played in the fourth century AD, rejecting the false and outdated belief that the Roman history of this region was irrelevant and unworthy of scholarly attention. Hopefully, researchers will be encouraged to explore this region and study the treasures that had been already uncovered but that often lie unstudied, if not sadly forgotten.1 This study, in turn, should motivate more exploration in Calabria, which without doubt has still much more to reveal about its Roman past.

1 See the Villa of Naniglio, for instance, which, since its discovery and publication in the 1960s, has been almost entirely neglected. 3

CHAPTER ONE: The Geography and a Brief Historical Survey of Calabria.

The Geography of Calabria (the Roman Bruttii)

Calabria is the narrow peninsula approximately 250 kilometers long at the southern extremity of Italy. It is located between the Tyrrhenian Sea to the West and the Ionian Sea and Gulf of to the East. The Strait of separates the region from Sicily. In classical times ‘Calabria’ denoted what is now eastern Puglia, the modern Salento, whereas modern Calabria in ancient sources was called with the plural Greek form Brettii, or the plural form of Bruttii. This was the name of the people that inhabited the region. The term Bruttium is often used in academic literature to refer to modern Calabria, but classical texts used the ethnic name to refer to the region.2 Framed by 800 kilometers of sea coast, nearly 49 % of Calabria’s land is hilly, 42% mountainous and only 9% plain.3 From north to south lie the ranges of Pollino, Sila, Serre and Aspromonte. The main plains in the region include that of Sibaris in the north, Lametia in the center, and in the south. The Pollino Mountains in the north are rugged and form a natural barrier separating Calabria from and the rest of Italy. This area is heavily wooded, while some parts are vast open plains with little vegetation. The Sila is a mountainous and forested plateau, about 1,200 m. above sea level with the highest point reaching 1,928 m., stretching for nearly 2,000 square kilometers along the central part of Calabria. The area boasts numerous lakes and dense coniferous forests. In the northern part, the Sila borders with the valley of the River (the longest in Calabria, at 580 m.) and the Plain. The Romans named ‘Sila’ the mountains located in the South of the region, the modern ‘Aspromonte,’ rather than the mountains of the northern region. The Aspromonte ridge stretches down to Calabria’s southernmost tip and is bordered by the sea on three sides. Its highest point is at 1,995 metres, and is full of wide, man-made terraces that slope down towards the sea. Another southern group of mountains is the ‘Serre.’

2 MAZZARINO, S., “Si puó dire ‘Bruttium’? La denominazione tardo romana dell’attuale Calabria, in Archeologia Storica Pugliese,” 25 1972, pp.463-467. Calabria acquired its present name in the time of the Byzantine domination, after the region had fallen into the hands of the Longobards and was lost by the Eastern Empire about 668 AD. The name Calabria, used in the modern sense, first appeared in Paulus Diaconus's Historia Langobardorum (end of the eighth century AD). 3 A detailed geographical description of the region is found in , A., Storia della Calabria, 1994; See also, GAMBI, L., La Calabria, Torino 1965. 4

1.1. Regional Map of Calabria (ancient Bruttii).4

Since the mountains of Calabria are so near to the sea, the rivers are all very short. They are called ‘fiumare,’ meaning short rivers. They stay dry for a long part of the year. During antiquity, the region was crossed by navigable rivers described by Pliny: "omnes ibi navigabiles Carcinus (Corace), Crotalus (Alli), Semirus (Simeri), Arogas (Crocchio), Thargines (Tacina). Those rivers disappeared due to a devastating deforestation that started with the Roman occupation, compromising forever the hydrological system of the region.5 In many respects Calabria is a land of dramatic contrasts. No other region in Italy features such a range of scenery and vegetation in a short space: high mountains rising from rugged coasts, valleys and plains. The climate is typically Mediterranean with below zero wintertime temperatures in the mountains and summertime temperatures reaching over 40°C along low valley areas.6

4 After http://www.kalabrien.biz/cartina.html 5 See “Paesaggi tardoantichi: alcuni problemi”, in FRANCOVICH, R.-NOYÉ, G., (eds.), Atti del Convegno. La storia dell’Alto Medioevo italiano (VI-IX secolo) alla luce dell’archeologia (Siena 1992), Firenze 1994, pp. 85-98. 6 Very likely, the climate in antiquity was different from what it is today. For a treatment on the climatic and ecological diversity within the Mediterranean world, See HORDEN-PURCELL 2000. 5

A brief survey of Calabria’s ancient history

The geography of Calabria accommodated several civilizations throughout history. Beginning in the eighth century BC, Calabria was colonized by the Greeks, who settled heavily along the coast. Rhegion, , Kroton, Lokris, as the Greeks called them, were numbered among the leading cities of during the sixth and fifth centuries BC. The Greeks were conquered in the third century BC by roving Oscan tribes from the north, including a branch of the Samnites called the Lucanians and an offshoot of the Lucanians called by the Greeks Brettii. The Brettii occupied the southern part of the modern Calabria. According to , they were an offshoot of the Lucanians (Lucani), for whom they worked as shepherds.7 The Brettii established the main cities of Calabria, including Consentia, the modern capital of Cosenza. Around 356 BC the Brettii rebelled against the Lucanians and split from them. Perhaps it was Dionysius II of Syracuse who solicited the Brettii to rebel against the Lucanians and used the Brettii to put pressure on the Greek colonies of Calabria, particularly Lokris, which were his allies. Afterwards, the Brettii created their own league that had its capital in Consentia (Cosenza). In the following years, they fought several Greek cities along the coast, including Hipponion (the future Roman colony of ). The relationship between the Brettii and was confrontational. The Brettii became allies of Pyrrhus and sided with against the Romans during the . Later, in 132 BC, they were punished by Rome, which broke their league, confiscated their territory, and founded numerous colonies, including Croton, Copia (in the territory of Sibaris and ), and Vibo Valentia (the Greek Hipponion). Croton and Tempsa were the first Roman colonies founded between 194 and 192 BC. 8 Dionysius of Halycarnassus (ca. 60-7 BC) accounts for the heavy exploitation of the forests in Bruttii by the Romans.9 During the time of the foundation of the colonies, a Roman road called Via Popilia was built. This connected Capua with Rhegium and became an important artery for military, political and economic reasons. It corresponds approximately to the modern highway Salerno-Reggio Calabria. Part of the Via Popilia is engraved on the Lapis Pollae (Polla Tablet).10 There were also two costal roads: 1) the Tyrrhenian way departed from , reached and joined the Via Popilia north of Vibo Valentia; 2) the Ionian way connected Rhegium with and Tarentum along a course traced by the Greeks, which passed by , Croton and Copia. A series of less important isthmian ways are also attested. They linked the Tyrrhenian and Ionian coasts with the hinterland. In addition to the major roads, the region had several important ports including Thurii-Copia, Vibo, Croton, Scolacium ,Velia, Locris, and Rhegium all favoring maritime communication and trade with the rest of the Roman world.

7 Strabo, Geography, 6,1, 4. 8 SANGINETO 1994, p.563 with a rich bibliography in the notes. 9 JACOBY, C., Dionysii Halicarnasei Antiquitatum romanarum quae supersunt, IV, Lipsiae 1905; CANTARELLI, F., Dionisio di Alicarnasso. Storia di Roma arcaica (Le antichità romane), Milano 1984. 10 S. Pietro di Polla, north of Diano Valley, see GORDON, A. E., Illustrated Introduction to Latin Epigraphy, 1983, 12, pp. 87-89; pl. 8. 6

1.2. The Regio Tertia: et Bruttii.11

From the time of , Bruttii together with Lucania (modern Basilicata) formed the Regio Tertia (fig.1.2). Later, under the emperor Diocletian this became an autonomous province called Lucania et Bruttii, which lasted until the Byzantine domination.12 The province was bordered by the river Bradanus (Bradano), which separated it from Apulia at the north-east. The river Silarus (Sele) separated it from at the north-west. On the Tyrrhenian coast the border between Lucania and Bruttii, according to the ancient sources was the river Laus (modern

11After http://upload.wikimedia.org/wikipedia/commons/d/d0/Shepherd-c-030-031.jpg 12 The Diocletianic province of Lucania et Bruttii in comparison to the Augustan Regio Tertia did not keep Metapontum but acquires the region north of the river Sele, including Salernum. See GRELLE, F., “Ordinamento provinciale e organizzazione locale nell’Italia meridionale, in L’Italia Meridionale in età’ tardoantica, Atti del Trentesimo Convegno di Studi sulla Magna Grecia, Taranto 2-6 ottobre 1998, p.120. 7

Lao). The border on the Ionian coast is not well defined in ancient sources. Perhaps it was located between the towns of Metapontum and Sibaris on the Gulf of Tarentum. The economy of Bruttii in Roman times was mainly agrarian.13 Roman ancient sources speak of specialized cultivations together with animal rearing and forest exploitation. For instance, Varro writes about sheep and cattle rearing (“Nobiles pecuariae in Bruttiis habentur (De rustica, II, 1, 2), and agricultural remunerative production, “In Italia in subaritano agro dicunt cum centesimo redire solitum” (56, 29). (Tull., 6-9; Att., III, 2) mentions the existence of several fundi situated in Bruttium. Pliny wrote that wine was produced in Thurii (NH, 14, 3, 39; 14, 8, 89), and that the wine of Consentia and Tempsa during his time was becoming popular (NH, 14, 69). Strabo wrote about the wine of Thurii (VI, 1, 14), and mentions the growing meadows of Vibo Valentia (VI, I, 5). Varro referred to a double crop of apples in the area of Consentia (De rustica, I, 7, 6), and Pliny mentions a triple (NH, 16, 115). Based on the sources, it appears that in Bruttii breeders and shepherds were also farmers and wine producers according to Varro’s account that a dominus should be both a farmer and a breeder in order to make the best profit (De rustica, 2; praef. 5,1,2,21).14 Archaeological data show that the Romans exploited the same areas that continue to yield agricultural products even today. These are mainly the lowest slopes of Calabria together with the Tyrrhenian and Ionian coasts, which have been agricultural for centuries, including the towns of Copia-Thurii, Sibaris, Croton, Scolacium, , Locris, Rhegium and Vibo Valentia. For centuries the coastal soil was fertile and used for viticulture, orchard trees and plantation of olive trees. The mountainous areas were used by transhumant cattle, and were largely exploited for the extraction of timber and pitch. The intensive deforestation promoted by the Romans caused the growth of marshland along the coasts. Yet, the prevalent picture of Bruttii provided by ancient authors ranging from the late Republic to the early imperial times is that of a deserted region infested by malaria, inhabited by shepherds and rude woodmen. 15 On the contrary, the high number of villas spanning from the first century BC to the fourth-fifth century AD located mainly on the coasts, prove that Bruttii provided profitable resources prompting the entrepreneurial elite to invest their fortunes in this region throughout the Roman occupation. Archaeological evidence shows that Bruttii underwent a revival of the economy especially during the later imperial times, participating in a general phenomenon of prosperity of all Southern Italy.16 In the last 15 years, more extensive and scientific excavations with the discovery of several late imperial villas in the region have motivated a new interest in this phase of Roman Bruttii, and a reevaluation of the information provided by the ancient sources. Until the 1960s, scholars did not pay much attention to late-antique Bruttii.17 Before then, researchers focused mainly on the prehistoric, and especially the Greek culture of Bruttii, which was commonly considered the highest form of civilization reached by the region. On the contrary, the Roman phases, particularly the late antique ones, did not receive much attention by researchers. This

13 SANGINETO 1994. 14 Ibidem. 15 Against this opinion of Calabria’s landscape during Roman times see KUZISCIN, V. I., L’organizzazione del latifondo in Italia alla fine della repubblica, in CAPOGROSSI COLOGNESI, L. (ed.), L’agricoltura romana, Roma 1981, pp. 15 ff. 16 Sangineto catalogues 22 villas in Bruttii dated to the late Roman centuries. See SANGINETO 1994, pp.569 ff. 17 Among the main contributions, see CRACCO RUGGINI 1961; KAHRSTEDT 1960; ARSLAN 1974 and 1990; on Scolacium, see ARSLAN 1991; COLICELLI 1995; SPADEA, R., Scolacium: una città romana in Calabria: il museo e il parco archeologico, Milano 2005. 8

may have been a consequence of the traditional view, handed down from the medieval literary tradition, that Bruttii reached its apex with the Greek civilization, but underwent an irreparable impoverishment and continuous decline in Roman times. In 1989, a conference held in Rome on the Late Antiquity of Bruttii contributed to a better understanding of the archeological sites of this region during the late Empire. 18 That same year, P. Arthur published an article on the production of Keay 52 amphorae. This scholar maintained that these amphorae were produced in Bruttii between the fourth and seventh century AD and used to export wine to the northwestern area of the Mediterranean from to Marsiles. 19 Another remarkable datum is the abundant presence of African ceramics and amphorae in Bruttii ranging from the fourth century until the eighth century AD. This fact points to intense trade with , probably with the mediation of Sicily. Africa during the latest Roman centuries had the strongest and most successful economy of the Empire, which might have contributed to the wealth that Villa at Palazzi di Casignana displayed in the fourth century AD.20 Between the fourth and fifth centuries AD many country settlements gradually became villages or monasteries, which at times survived to the fall of the Empire and the invasions of the Barbarians.21 The economic vitality of Calabria must have continued if at the beginning of the sixth century AD Cassiodorus still praised the production of grain, oil, wine, vegetables, cheese, meat, wood and pix of the region.22 The Late Antiquity of Calabria in the last few decades has been the object of careful examination by scholars such as G. Fiaccadori, and G. Noyé.23 Their studies have continued to expand our knowledge of Calabria during Late Antiquity encouraging new generations of scholars to carry on their work. Moreover, the most recent archeological discoveries led by the Superintendence of the Archaeological Treasures of Calabria, keep shedding a light on the later Roman history of Calabria.24 This, in turn, compels historians to provide an explanation for the existence of lavish villas that were built during the late imperial period such as the Villa at Palazzi di Casignana. This complex so far stands as the most outstanding proof that the Roman phases, particularly the later ones, of Calabria were indeed as worthy of note as the Greek past of the region.

18 SANGINETO 1994, p. 592. 19 ARTHUR, P., “Some observations on the economy of Bruttium under the later Roman empire,” in Journal of Roman Archaeology, 2, 1989, pp.132-142. 20 SANGINETO 1994, p. 592. For an account on African ceramics and amphorae imported in Bruttii, see “Atti del Convegno dell’Ecole Française of Rome,” in Melanges Ecole Française of Rome (M), 100, 1991, 1. A large number of African ceramics and amphorae discovered in Scolacium are discussed by RACHELI, A., “Le ceramiche da mensa, le anfore”, in R. SPADEA (ed.), Da Skylletion a Scolacium, Roma 1989, pp.147-163. 21 ARTHUR 2004. 22 Cassiodorus, Variae, I, 16,-22; VIII, 31; XI, 39; XII, 4, 12, and 15. 23FIACCADORI 1994; NOYÉ 1996, 1998, and 2002. 24 Often the archaeological discoveries are not followed by exhaustive publications. New data is frequently publicized in the form of preliminary reports. 9

CHAPTER TWO: The Layout of the Villa at Palazzi di Casignana and the Archaeological Area.

2.1. General plan of the villa.25

25 After, SABBIONE, C. (ed.), La villa romana di Palazzi di Palazzi di Casignana, Gioiosa Jonica, 2007, p.32, fig.23. In this study, I will refer to this book as ‘Guida’. 10

Introduction

Villa at Palazzi di Casignana provides an outstanding example of lavish architecture in the late Empire (fig.2.1). The areas of the villa so far uncovered are limited in comparison to the original extension of the site in antiquity. The villa was originally built in the first century AD and endured nine phases of transformations, renovations, and additions.26 Visible today are the structures of the last phase dated to the fourth century AD when the villa reached its highest splendour and maximum expansion. The building was abandoned around the first half of the fifth century AD for unknown reasons, but on the same site developed a small human settlement that continued to exist until the seventh century AD. To date, Villa at Palazzi di Casignana stands as the best known late Roman villa of Bruttii, having been largely excavated and published.

2.2. Aerial view of the villa.27

The figure above (2.2) gives an idea of the grandiosity of the complex and the distribution of space. Approaching the villa from the north , one is immediately struck by the large span of the property which covers 5000 square meter. The villa is perpendicular to the seashore and is orientated west-east. Its location was strategically chosen in accordance with the precepts of Roman agronomists. It was built in a place that was salubrious all year long.28 The villa faced eastward into the rising sun and healthy winds.29 It rested on a large fertile plain at the foot of

26 For a detailed account of the previous phases of the complex see Guida, passim. 27 After Guida, p.36, fig.29. 28 Varro, I, 4, 3-4: ‘Utilissimus autem is ager qui salubrior est quam alii, quod ibi fructus certus…Quare ubi salubritas non est, cultura non aliud est atque alea domini vitae ac rei familiaris’;Columella, I,4: ‘Quod incohatur aedificium, sicut salubri regione ita saluberrima parte regionis debet constitui…’ 29 Columella, I, 5: ‘Optime autem salubribus, ut divi, locis ad oriente vel ad meridiem, gravibus ad septentrionem villa convertitur’. 11 hills,30 close to the River Bonamico and 300 m. from the scenic Ionian shore.31 Particularly suitable for the Villa at Palazzi di Casignana was the abundance of water likely supplied by the nearby river Bonamico and perhaps by means of an aqueduct.32 An important factor in the choice of the site must also have been its proximity to the main Roman road alongside the Ionian coast.33 This road stretched from Rhegium to Metapontum and Tarentum on the same route created by the Greeks. It facilitated communication with the coastal towns, including Locri (which is about 15 km. distant) Croton and Copia complying with the instructions of ancient agronomists.34 Located about 80 km. away from the villa was Rhegium, one of the most important ports of the Mediterranean Sea. These urban centres likely provided their services, markets and harbours to the villa owners and the rural population of Palazzi di Casignana. Another important road was the internal Via Popilia connecting Rhegium to Capua. This also favoured travel and trade between Bruttii, Central Italy and Rome. It is even possible that the villa had its own harbour as indicated by massive walls recently discovered in the mouth of river Bonamico.35 This river very likely was navigable in antiquity providing the villa with easy access to the hinterland.

30 Varro, I, 13,7: ‘Item videre oportet si est collis, nisi quid impedit, ut ibi potissimus ponatur villa’;Columella, I, 4: ‘Haec igitur est medii collis optima positio’. 31 Varro I, 16, 6: ‘Eundem fundum fructuosiorem faciunt vecturae, si viae sunt, qua palustra agi facile possint, aut flumina propinqua, qui navigari possint, quibus utriusque rebus evehi atque invehi multa ad praedia scimus’. The sea moved backwards from the shore but perhaps was closer to the villa in antiquity. 32 A pillar was uncovered on the villa’s site. It likely belonged to an aqueduct connected to River Bonamico, see Guida, p. 35. The closeness of the villa to water is prescribed by Roman agronomists. See Varro, I, 11, 2: ‘Villa aedificanda potissimum ut intra saepta villae habeat aquam, si non, quam proxime’; Columella I, 5: ‘Sit autem vel intra villam vel extrinsecus inductus fons perennis…’. 33 The Ionian coastal road has been partially uncovered, see GIVIGLIANO 1994, pp. 318 ff. The proximity of the estate to the sea, navigable rivers or important roads is recommended by ancient agronomists, see Cato, I, 3: ‘Praedium prope siet, aut mare, aut amnis qua naves ambulant, aut via bona celebrisque’; Varro, I, 16, 1: ‘…si viae aut fluvii, qua portetur, aut non sunt, aut idonei non sunt’; Columella, I, 3: ‘Post haec duo principalia subiungebat illa non minus intuenda: viam, aquam, vicinum. Multum conferre agris iter commodum: primum, quod est maximum, ipsam praesentiam domini, qui libentius commeaturus sit, si vexationem viae non reformidet; deinde ad invehenda et exportanda utensilia…’ 34 Cato, I, 3: ‘…oppidum validum prope siet.’ 35 Guida, p.35. 12

The Layout of the villa

North

2.3. Plan of the villa.36

36 After Guida, p.32, fig.29. 13

The villa is formed by two main areas: Area A and B, on the western side, and Area D and H on the eastern side (fig. 2.3). They are arranged around the internal Courtyard E (elsewhere cited as Courtyard 55) that is currently occupied in its northern side by a modern building that obliterates ancient structures. The courtyard measures 30 m. x 24 m. It can be compared to those at the Sicilian villas at Patti Marina (33.5 m. x 25 m.); Tellaro (20 m. x 20 m.); Piazza Armerina (40 m. x 18 m.); and the Villa at Desenzano on the Garda Lake shore in northern Italy (18.6 m. x 17.2 m.).37 It was not only a connecting area between several sectors of the villa, but likely also a gathering place linked to the baths.38 In antiquity, Courtyard E connected Area B with area D-H, but unfortunately the construction of a modern road cut off and disrupted the villa’s plan, separating the two sectors. Hence, the modern visitor coming from Area B must cross the modern road SS106 in order to visit Area D-H located in front of the seashore. Area B is occupied by two suites of baths. Area D-H likely had a residential and representative use. A brief description of the villa complex follows (see the villa’s plan on fig.2.3 for a graphic representation).

2.4. of the villa.39

Area A is located on the north-western side of villa. Very likely it was a garden that in large part was paved with gravel.40 At the furthest western side of Area A there is a monumental

37 Patti: WILSON 1990, p.200, n.5; Tellaro: WILSON 1990, pp.206-209; Piazza Armerina: WILSON 1983; Desenzano: SCAGLIARINI CORLAITA et Alii 1997. 38 The function of the courtyard is discussed in Chapter 4 of this study. 39 After Guida, p.62, fig.76. 14 fountain (nymphaeum, fig. 2.4) that was built during the last phase of the villa dated to the fourth century AD. The fountain underwent several phases of construction, which can be proven by visible portions of earlier structures that were covered by the newer fountain. The nymphaeum was formed by a rectangular tank with an apse in the background (the apse measures 5 x 5 m.), and had four corner pilasters that likely supported the roof of the fountain. On the front fountain there is a large base made with limestone blocks. On the corner blocks of the base are visible two round hollows where perhaps two columns stood providing the fountain with an elegant prospect. Behind the nymphaeum there are five rectangular water tanks. These may have been built to preserve water for the daily needs of the villa and its estate. Yet, it is not excluded that four of them were used to rear fish, a practice that was not uncommon in Bruttii during the Roman times.41

40 For a treatment on gardens in Roman country villas see LITTLEWOOD 1987. More recently, K. T. VON STACKELBERG, The Roman garden : space, sense, and society, New York, 2009. 41 See Chapter 8 of this study for a discussion on the water tanks at Palazzi di Casignana and their probable use. 15

North

2.5. Area B with two suites of baths: the Western Baths (in red) and the Eastern Baths (in orange).42

Area B (fig.2.5) is situated at the north-west side of the villa and contains the private baths consisting of two adjacent and communicating suites: the Western Baths (A, in red), and The Eastern Baths (B, in orange). The baths are exposed to the afternoon sun, in line with Vitruvius’ instructions.43 Both featured similar characteristics: a frigidarium, an unheated area with two tubs (Tubs 12 and 25), and a calidarium, a heated area formed by three adjacent rooms (24, 36, 29). The calidarium of the Eastern Baths is formed by Room 10, 11, and 37. In both baths, the frigidaria, Rooms 12 and 25 occupy the largest space, and are more monumental and richly decorated.

42 After Guida, p.38, fig. 33. 43 Vitruvius, De Architectura, V, 10.1. 16

2.6. Areas G, C, D, and H.

Area G (fig.2.6) is located at the north side of the courtyard and remains unexplored due to the presence of a modern building that was built on top of the ancient structures. Therefore, the use of this area of the villa remains unknown. Area C is at the south side of Courtyard E. It has been interpreted as a service area because the rooms of this sector are not particularly decorated and include two latrinae. It is located in a transitional area between the baths (Area A- B) and Area D-H. Area D and H (fig. 2.6) face the Ionian coast (at 300 meters distance) on the east side of the villa. This sector features a monumental façade with a columned portico stretching 22.50 meters and ending on the short sides with two apsidal rooms. Although only the ground level structures remain, V. de Nittis reconstructed these rooms featuring two towers that in fourth century AD would have flanked the portico, providing the villa with a fortified prospect by the sea.44 This reconstruction seems unlikely and will be discussed in more details in Chapter

44 See V. DE NITTIS, in Guida, pp.107-113. 17

4 of this study. Area D and H instigate the most intriguing questions in regards to the architecture, ownership and use of the villa. Area D contains the largest room of the residence, Room I (11.50 m. x 13.60 m.), which together with the monumental sea front façade signify the importance and representative role of this sector of the villa.45 The splendid grandeur that the Villa at Palazzi di Casignana evokes is even more astonishing when one walks over the lavish floor mosaics that decorate several rooms of the complex, and sees the remains of marble revetments that originally covered the walls up to the ceiling inside the most imposing rooms of the villa. So far, 23 floor mosaics have been uncovered and restored inside the villa, constituting the largest single collection of Roman mosaics in Calabria. Glass mosaics very likely decorated the vaulted ceilings of several rooms as indicated by coloured glass tesserae found scattered in the rubble of the fallen structures. The awe-inspiring vision of such a wealthy residence grows even more when one thinks of the long lasting belief, handed down to us from both ancient and modern literature, that Bruttii in Roman times was an economically insignificant, and impoverished region, especially in the latest centuries of the Roman occupation.46

45 When looking at the entire plan of the fourth century villa, one immediately notices the orderly distribution of space. The main axis of the villa is marked by important structures as Room 7 (likely used for reception), Courtyard E (the main peristyle), and Room I (the most monumental room of the entire complex). On the contrary, in other late antique buildings, such as the villa at Piazza Armerina, the main groupings of rooms are loosely related and located on differing axes. 46 Scholarly attention towards the Roman time of Bruttii started in 1960s with the study by KAHRSTEDT, U., Die wirtschaftliche Lage Grossgriechenlands in der Kasierzeit, Wiesbaden 1960. For an account of the early scholarship see PAOLETTI 1994, pp.471-474. 18

CHAPTER THREE: The Archaeological Area of the Villa.

3.1. Archaeological area of the Villa at Palazzi di Casignana.47

Introduction

Before stepping through a detailed description of each room, it is helpful to survey the area surrounding the Villa at Palazzi di Casignana to see how this building was the centre of a larger settlement. This chapter describes the archaeological findings near the villa including traces of roads, remains of several buildings, and two necropolises that lie at the north-east and south area of the villa (fig. 3.1). The presence of two large necropolises at the site during the late imperial phases calls for consideration. Burials are important indicators of territory occupation and land organization. Their placement near the villa while it was still inhabited, and even after its abandonment, invites us to study the funerary practice at Palazzi di Casignana more thoroughly. This chapter also discusses and challenges the hypothesis that the villa belonged to a statio, proposing instead that the villa was a private residence owned by a very wealthy Roman during the fourth century AD. As new discoveries are made at the site, archaeologists can provide a better understanding of the true function of this complex, which today largely remains unknown.

47 After Guida, p.35, fig. 28. 19

The archaeological area of the villa

West Perpendicular to the nimphaeum of the villa and its water reservoirs

North

3.2. Plan of the villa. extended a road running South-West/North-East. This road ended in a large open space in front of the villa’s Portico 17(See arrow near area A, fig.3.2). On the road stood a building formed by two adjoining rooms built with small stones. The front of this building features a portico, the foundations of which still remain in situ. The function of this small structure is unknown.

20

Southwest Ruins from other buildings were found scattered in the area, including a large rectangular building that lies 150 m. southwest to the villa. Near this building a pilaster of bricks was uncovered, which probably belonged to an ancient aqueduct that furnished water to the villa and its baths.

South The southern side of the villa was lined by a strip of land progressively occupied by several buildings. Beyond this line, a building with four rooms and a large apsidal space was erected with the same orientation of the villa (fig. 3.2., area F in green). Only the foundations remain of this building without traces of pavements. In the southern area of the villa other scattered structures have been identified. They are spread all the way to the area where a modern tourist farm house is located (see Agriturismo, fig. 3.1). East of the Agriturismo, a necropolis was uncovered (fig. 3.1). It contained graves where the dead are laid on tiles or directly on the ground and covered with roof pitched tiles (“tombe alla cappuccina”). These burials have been vaguely dated between the imperial time and the .48

North The villa was skirted by a second road (4.50 m. wide) running east-west, which also ended in a large open area located in front of Portico 17 of the villa. In this area, portions of a monumental building were unearthed (Building G, in green, fig.3.2.). It originally measured 21 m. x 25.5 m. and featured three squared rooms connected by a portico with pilasters, opening onto the large outdoor area at the north side of the villa where the monumental nymphaeum stood. The regularity of the plan, the location by the main roads and the thick walls (90 cm.) of building G have suggested to the villa’s archaeologists that this building served a public function.49 Although current excavations do not provide sufficient evidence to interpret the original function of Building G, it may be inferred that its use was closely connected to the baths. Building G may have been a place for exercising or preparing for the baths. Its portico with pilasters formed a sort of belvedere looking out to the garden and its monumental fountain. For this reason, it would have been a suitable and pleasant place for walking or exercising for the bathers before they moved into the baths. Scattered remains of several buildings have been found underneath a modern camping area (Campeggio, fig.3.1.), in the northern side of the villa. They are located along the modern road SS 106 and face the seashore. The discovery of several outbuildings around the Villa at Palazzi di Casignana is of great importance. A careful examination of them may reveal a great deal about the economic basis of this villa and its estate.

The Necropolises

The presence of two graveyards found at the north-east and south area of the villa (fig. 3.1) is also quite significant, especially with additional tombs uncovered more recently near the eastern area of the villa where Room I stands.50 The location of the north-eastern necropolis may

48 Guida, pp.89-93. 49 Guida, p.34. 50 This is the largest and most monumental room of the complex that had a cross-shaped plan. I have personally seen the tombs as they were being uncovered near the eastern façade of the villa. They do not appear in the archaeological report La Villa Romana di Palazzi di Casignana. Guida Archeologica, (2007), because they were discovered after the publication of the book. 21 have been initially determined by the existence of a road that ran nearby. Both necropolises at the Villa at Palazzi di Casignana have been dated somewhere between the third and fourth century AD, making them contemporary with the latest phases of the villa.51 This implies that the dead were buried near the villa not only while it was occupied but also during a period when the villa was renovated into a monumental estate. The graves at Palazzi di Casignana may disclose a great deal of information about the social status, rank in life, gender, way of living and the religious beliefs of the people in the villa’s territory. A careful study of the dead, their grave goods together with the placement and orientation of the bodies may reveal a better picture of the social organization of the Villa at Palazzi di Casignana during the late imperial phases. In general, the characteristics of burials are little known because they were destroyed by agricultural work or removed. Especially in past excavations, when proper scientific methods of recording were not systemically applied, burials found near late Roman villas were treated as less important findings in comparison to the architectural structures that came to surface in their proximity, especially when these were aesthetically attractive. As a consequence, Roman ways of burial have received relatively scant attention, even if the Romans have left abundant material evidence about their funerary rituals (more work has been done on the Greek and Etruscan ‘archaeology of death’). In modern archaeological surveys graves have proven to be important connections to the Roman way of thinking and religious beliefs. Recent scholarship is paying greater attention to the phenomenon of burials located near villas.52 Burials can tell us a great deal about the social status of the dead, the territorial occupation of any particular area and how the dead were commemorated by a certain group of people. The presence of the necropolises near the Villa at Palazzi di Casignana and their significance is discussed on Chapter 9 of this study.

Villa at Palazzi di Casignana as the nucleus of a statio

The luxurious character of Villa at Palazzi di Casignana and the discovery of buildings, traces of roads and burials around the villa has brought scholars to hypothesize that the villa was a public facility, namely a statio, one of the many travel posts created along the Ionian coast by the Romans.53 This hypothesis is based on information provided by Roman itineraria,54 which show a series of stationes located at an interval of about 20 Roman miles (30 Km.) on the Roman Ionian road that went from Rhegium to Locri and beyond.55 Literary evidence of stationes along the Ionian coast certainly suggests that the villa at Palazzi di Casignana served as a statio. Yet, the discrepancies of the ancient sources that record

51 For a detailed description of the two necropolises see Guida, pp. 89-90. 52 ARCE 1997 b, LEWIT 1991. 53 CORSI 2000, p.131. 54 Roman itineraria were lists of the for travellers. The Itinerarium Antonini is the most ancient and complete register of the stationes and distances along the various roads of the Roman empire, containing directions how to get from one Roman settlement to another. It includes an appendix on the major maritime routes of the Mediterranean Sea. The Itinerarium Antonini is seemingly based on official documents, probably of the survey organized by , and carried out under Augustus. Due to the scarcity of other extant sources of this information, it is a very valuable source. Nothing is known with certainty as to the date or author. It has been argued recently that the Itinerarium Antonini was compiled in the second quarter of the fourth century, though largely on the basis of third century sources. See ARNAUD, P., “L’Itinéraire d’Antonin: un témoin de la littérature itinéraire du Bas-Empire,” Geographia Antiqua 2 (1993), pp.33-49, especially pp.43 ff. See also, CORSI 2000, p.60. 55 GIVIGLIANO 1994, pp. 318-324. This tract of the Ionian Roman road may have corresponded approximately to the modern road SS 106. 22 stationes, the lack of significant archaeological evidence of these facilities across the Empire, and negative comments about the cursus publicus contained in fourth century texts induce to contradicting this hypothesis. Libanius in his letters and orations laments the disservice of the cursus publicus during his time. He writes that it was inefficient and poorly organized and that only with the emperor Julian (AD 332-AD 363) was the system finally improved.56 Public structures at this time not only decrease in number but also in lavishness as the municipal government does not invest in them as before.57It is more likely that the Villa at Palazzi di Casignana was a residence or the pars urbana of a private villa, and the central nucleus of a larger architectural complex that included facilities where agricultural and industrial activity occurred supported by peasants living in smaller dwellings with close proximity to the villa. In late Roman villa architecture, the separation of the pars urbana from the pars fructuaria- rustica, which was devoted to the agricultural and crafts activities, is frequently attested.58 Although no pars fructuaria has been yet identified at Palazzi di Casignana, the discovery inside the villa of a large amount of wine amphorae Keay 52 produced in Bruttii strongly indicate that the villa’s estate may have been involved in viticulture and wine commercialization during the latest phase of its occupation. Wine production and exportation was strong in Bruttii during fourth century AD, an activity that will be discussed in Chapter 8 of this study. Moreover, Villa at Palazzi di Casignana’s long inhabitation from the first to the mid-fifth century AD, and especially the care and investment in remodelling and enhancing it over the centuries, strongly indicate that the villa was a private residence handed down from generation to generation, although not necessarily belonging to the same family. It is possible that the villa passed to a different ownership when it underwent the monumental renovation in the fourth century AD. Several aristocratic and monumental residences dated to the fourth century have been discovered in many regions of the Empire showing that private wealth reached a very high level competing with public and imperial properties. The grand scale of the Villa at Palazzi di Casignana may well have been the result of private wealth in the fourth century, since the Roman elite of Bruttii during that period were likely involved in a successful economy. The hypothesis that Villa at Palazzi di Casignana was the nucleus of a statio named Altanum, situated on the Ionian coast south of Locri, is mainly based on the

56 DI PAOLA 1999, pp.99-103. 57 JONES 1974, pp.980-985. 58 On this topic, see Chapter 8 of this study. 23

3.2.Synoptic picture of the Itineraria.59

59 After GIVIGLIANO 1994, p.319, fig.53. 24

Itinerarium Provinciarum Antonini Augusti.60 This document locates the statio of Altanum 20 Roman miles (30 Km.),61 after the statio of Subsicivo near Locri.62 But Villa at Palazzi di Casignana is located 15 km south of Locri, not 30 km. This is one first argument against associating the Villa at Palazzi di Casignana with the statio of Altanum. A maritime itinerarium called the Itinerarium Antonini Maritimum lists a statio called before Rhegium at the same point where the Itinerarium Antonini records the statio of Altanum (see fig.3.2). It has been suggested that the name Altanum of the Itinerarium Antonini refers to a ‘western wind’ and would be the Latin translation of Zephyrium, the statio recorded on the Itinerarium Maritimum. Thus, the names Altanum and Zephyrium recorded by the two itineraria most likely refer to the same statio.63 Givigliano suggests that Altanum could correspond to the modern ‘Capo Bruzzano,’ a promontory on the Ionian coast exposed to the wind situated at about 8 km

3.3.Capo Bruzzano and the Tyrrhenian coast near Locri.64

south of Palazzi di Casignana (fig.3.3). According to Givigliano, it is also possible that Altanum was situated on a completely different location. He suggests that Altanum was on an isthmian road crossing the Aspromonte Mountains, which connected the Ionian coast with the Tyrrhenian (fig.3.4). This isthmian road likely started 30 Km. (20 Roman miles) beyond

60 For a treatment on this Itinerarium see GIVIGLIANO 1994. The problem of the exact identification of Roman posts along the Ionian coast from Rhegium to Caulon is discussed by M. Cardosa. See Guida, pp.19-21. 61 The average distance from one statio to the other ranged was between 30 and 36 km. The same distance according to Vegetius (I, 9) was normally covered in one day by a Roman legionary. 62The Latin name ‘Subsicivo’ would refer to an area that was ‘left out’ of the Roman repartition of Locri’ territory. For a discussion on this statio. GIVIGLIANO 1994, p.358, note 387. 63 Ibidem. 64 After Guida, p.20, fig.7. 25

Subsicivo, the statio near Locri. Givigliano’s assumption is based on ancient texts that record a place named Altanum on the plain of Piano Casciano at 700 m. height on the Aspromonte in direction of the Tyrrhenian coast. An abundance of Roman archaeological material was discovered in this area, and much of it can be seen today at the Museum of Polisena.

3.4.The Isthmian way between the Ionian and Tyrrhenian coasts.65

Identifying Altanum as a Tyrrhenian statio is convincing since it is supported by both literary and archaeological evidence. If Altanum is the statio located on the Aspromonte Mountains then Altanum and Zephyrium, as recorded by the two itineraria, would represent stationes located on the western Tyrrhenian coast, not the eastern Ionian coast. In contrast, C. Sabbione has proposed that Altanum of the Itinerarium Antonini corresponds to the site of the Villa at Palazzi di Casignana, and Subsicivo to a statio near the modern Marina di Gioiosa rather than to a statio near Locri. In order to follow the distance in Roman miles recorded by the Itinerarium Antonini, Sabbione proposes to switch the position of the statio of Cocinto with that of Succeiano.66 The scholarly debate on the location of Altanum shows how difficult it is from an archaeological standpoint to situate with precision Roman stationes by focusing on information provided by the Roman itineraria. These texts are valuable sources because they record names of stationes and distances between them, but their notable disagreements, which at times occur even within the same document, make it difficult for archaeologists to locate places with precision. The incongruity of the itineraria, according to C. Corsi, may be attributed to the

65 After GIVIGLIANO 1994, p.323, fig.61. 66 Guida, p.20. Before the discovery of Villa at Palazzi di Casignana the statio of Altanum was located near the town of , See CORSI 2000, p. 131. 26 different dates of the texts that were used to compile them, which reflect the continuous modifications and ‘mobility’ of the travel posts along the cursus publicus.67 In Italy, only in recent years has the study on the Roman road network expanded to include also the archaeological data moving beyond the analysis of literary and epigraphical texts. This new methodology is leading scholars to catalogue Roman travel posts more accurately.68 Stationes are typically associated with densely populated settlements. Yet, Roman itineraria and the Peutinger Table often record them with coordinates that do not correspond to inhabited centres. 69 According to Corsi, the current state of knowledge on the Roman network of roads prevents us to say if each place listed on the itineraria or the Peutinger Table is truly correspondent to a statio of the cursus publicus. The places listed on the itineraria may indicate localities that are not to be associated with stationes of the cursus publicus. This is because, according to Corsi, Roman itineraria may be private documents rather than lists of travel posts on the cursus publicus created by the State. It is possible that the itineraria record accommodations built by private patrons along the line of the imperial cursus publicus. Taking Corsi’s arguments into account, even if Villa at Palazzi di Casignana were a statio, it may have not belonged to the State but rather to a private patron. The fact that a statio was recorded by the itineraria does not automatically imply that it was a public structure. The Villa at Palazzi di Casignana may well have been a private residence that benefited from a strategic location on a major Roman road where several stationes were established over a long period of time from Augustus onward. At the same time, it is not excluded that the villa may have constituted a driving force for the creation of a public statio in the area.70 Written texts that record stationes are innumerable.71 The most common terms used by ancient authors are mansio, mutatio, statio, deversorium, hospitium, caupona, stabulum, taberna, popina, cenatio, ganeum, gargustium, thermopolium, xenodochium. The Codex Theodosianus uses interchangeably the words statio, mansio (from manere=stay) and mutatio (from mutare=change the equipment).72As Corsi points out, it is very difficult for us to discern if these words refer to a physical place or to a temporal space. Generally, literary sources describe stationes as travel posts and mansiones as facilities that provide all of kinds of services for travelers. Thus, it would be more correct to talk about mansio rather than statio in the case of the Villa at Palazzi di Casignana, if this complex had some connection with either one facility. According to Corsi, mansiones typically included a variety of structures, such as a guard post, rooms for the travellers, baths, storage rooms, barns, stables, workshops for horseshoes and cauponae. They had to supply fresh horses, mules, donkeys and oxen, as well as carts, food, supplies, and accommodation to soldiers, magistrates and messengers who travelled through the Empire. Mansiones were built by the State at major roadway interconnections and along important maritime ways in order to provide messengers and magistrates with commodities that

67 CORSI 2000, p.64. 68 CORSI 2000, p.1 ; see especially Chapter 1: Storia degli Studi. 69 For an account of the Peutinger Table see MILLER 1962. 70 The view that a Roman villa may develop into a statio is expressed by CORSI 2000, p.131. 71 Ibidem, pp.20 ff. for a discussion on the ancient sources and use of words related to the cursus publicus.

72 The name of a statio is based on: 1) the distance such as ad talem lapidem (to such mile stone); 2) a particular topographical feature as ad fines (to the border), ad fluvium (to the river), ad pontem (to the bridge);3) to a name of a town, although often stationes are at some distant from the town. See DI PAOLA, L., Viaggi, Trasporti e istituzioni. Studi sul cursus publicus, 1999. 27 facilitated the development of the cursus publicus.73 Mansiones were located at a regular interval of circa 30-36 km, a distance covered in one day.74 Moreover, it is not always clear from the texts what were the travel posts owned by the state for accommodating ambassadors, magistrates, and all kinds of public functionaries of the imperial court, and what were instead private and used by ordinary citizens. To complicate our understanding, the description of stationes has rarely found a correspondence in the archaeological data making the role of travel posts along the Roman roads very difficult to classify. In short, although itineraria or literary texts record the existence of several stationes on the Ionian road from Rhegium to Locri, they are not a sufficient proof to identify any statio with the location of the Villa at Palazzi di Casignana. Much more extensive archaeological explorations in the territory are required. Even if Altanum had to be located on the Ionian coast and in proximity of the villa, sources do not provide enough evidence to say that Villa at Palazzi di Casignana was the centre of a statio. The management and administration of mansiones (praepositura mansionum) in the fourth century AD was under the control of the State. To manage a mansio meant providing strong and robust animals as donkeys, oxen and horses (Codex Theodosianus 6, 29, 9, 412); and making sure that nobody used a number of animals higher than the one established by law (Codex Theodosianus 8, 5, 60, 400).75 At Palazzi di Casignana, except for the baths, there are no other structures that strongly correspond to particular components of a mansio. But baths were also a common feature in fourth century Roman villas, especially where the availability of space and the wealth of the owner made it possible even to double their number.76 Therefore, the baths at Palazzi di Casignana do not represent a strong indicator that this complex belonged to a statio. No stables or any buildings designed for receiving travelers of all ranks have been uncovered so far, or any structure that may point to a use of this villa as a travel post. It is more likely that the Villa at Palazzi di Casignana was privately owned during the fourth century AD.

73 The cursus publicus was created by Augustus to transport messages, officials, and tax revenues from one province to another. The service was still fully functioning in the first half of the sixth century in the Eastern Roman Empire, when the historian charges the emperor Justinian with the dismantlement of most of its sections, with the exception of the route leading to the Persian border. The term cursus publicus is established in the fourth century AD, even if according to Victoris it appeared for the first time under . More common in earlier times were the terms cursus vehicularis, vehiculatio and res vehicularia.” See DI PAOLA 1999, p. 21 with notes. About the distance between stationes: carts could travel about 8 miles a day, pedestrians a little more, and so each mansion was about 20 miles from the next one”. See also CORSI 2000. 74In addition to mansiones, mutationes were placed every 8-9 Roman miles. Mutationes were used merely to change horses or to take refreshment. There were four mutationes for every mansio.The Itinerarium a Burdigala Hierusalem usque, which is a road-book drawn up about the time of Constantine, mentions in order the mansiones from Bourdeaux to Jerusalem with the intervening mutationes, and other more considerable places, which are called either civitates, vici, or castella. The number of leagues (leugae) or of miles between one place and another is also reported. 75 A diplomata or certificate was issued by the emperor himself and was necessary to use the roads and the travel facilities. Governors and minor magistrates used the diplomata to give themselves and their families free transport abusing the system. Forgeries and stolen diplomata were also used. See DI PAOLA 1999; GRAHAM, M.W., News and frontier consciousness in the late Roman Empire, 2006. 76 See discussion on late Roman baths in Chapter 4 of this study. 28

The ownership of the Villa at Palazzi di Casignana

Archaeological data and ancient sources have not yielded any evidence in regards to the ownership of the Villa at Palazzi di Casignana. It is possible that the fourth century owner of the Villa at Casignana was a distinguished member of the new senatorial class formed under the emperor Constantine. The name “Palazzi di Casignana” of the modern municipality to which this villa belongs, may have derived from ‘palatium,’ namely a luxurious residence, and the name ‘Casignana’ from ‘Casinius’ a Roman landlord. Perhaps the palatium in question was the fourth century Roman Villa at Palazzi di Casignana, and Casinius its late imperial owner.77 Similarly, the name of the town Desenzano on the shore of Lake Garda in northern Italy, where another important fourth century villa was built, may have derived from Decentianum namely a fundus that belonged to a Roman landlord named Decentius.78 The two modern towns of Desenzano and Palazzi di Casignana would then represent the final development of a settlement that began with a suburban Roman villa and its estate. Whoever commissioned the renovation of the Villa at Palazzi di Casignana in the fourth century AD, very likely had economic and political interests in the area of Locri and throughout the region of Bruttii. The villa may have been a retreat residence of a fourth century aristocrat but also a seat of political and social gatherings where influential and prominent personages met during the late Empire. The dominus of Villa at Casignana, and the owners of other villas in the territory of Locri, may have been members of the senatorial class, perhaps natives of Bruttii, but also from the great aristocratic families of Rome and other regions of the Empire. Constantine created a new senate in Constantinople and increased the number of senators from 600 to 2000 recruiting them from the equestrian order and from the municipal nobility of the provinces and Italy.79 The senators of the provinces were typically natives of those provinces and owned estates near the municipia where they held political positions.80 Constantine supported the tendency of provincial senators to reside in their native territories.81 These aristocrats sought prestigious political positions in the regions where they owned estates in order to exercise patronage over their community.82 The same can be inferred about the owner of the Villa at Palazzi di Casignana.

77 BARILLARO 1979. 78 MIRABELLA ROBERTI in GHISLANZONI 1962. 79 Over time the recruitment of senators came to include the lower levels of the provincial governors. See HEATHER 1998, p. 190; NOVAK 1979. 80 HEATHER 1998, pp. 204-209. 81 On this subject see CHASTAGNOL 1977. 82 On the role of the senate in late antiquity see NOVAK 1979; GARBARINO 1988; MARCONE 1992-1993; ARHNEIM 1972; BARNISH 1988; CHASTAGNOL 1982 and 1992; CRACCO RUGGINI 1998. On patronage, see SALLER 1982. 29

CHAPTER FOUR: The Monumentality Phase of The Villa at Palazzi di Casignana in the Fourth Century AD.

Between the third and the fourth century AD, Villa at Palazzi di Casignana was expanded and enhanced with new structures. Figure 4.1 illustrates the modificiations that occurred during this time.83 The villa was embellished with the following features: 1) Two sets of monumental baths: the Western and the Eastern Baths, located at the western area of villa (A-B on fig.4.2). 2) Portico 17 was built at the western entrance of the villa (fig. 4.1 and 4.2). 3) Room 7 of the baths (fig.4.2) was made more elegant with the addition of a column at the entrance, three steps, and with a beautiful geometric mosaic on the floor. 4) The villa’s southern area, which perhaps was used for service, became larger. A long columned portico was added (Portico 59, fig. 4.3), enhancing the southern side of the villa’s peristyle. This portico connected the villa’s western area with the seafront sector. Several rooms of the southern area were remodeled. Latrina 40 was expanded and lavishly paved with marble slabs. 5) The villa’s seaside area became remarkably monumental. The largest and most lavish room of the entire villa, Room I, was built at this time in a central and prominent position. It featured a cruciform and apsidal plan, perhaps a vaulted glittering ceiling, an elaborate geometric mosaic and marble veneers. At the east of Room I, new rooms were built. They were decorated with fine polychrome mosaics. At the west, the rooms that were built during a previous third century AD phase underwent significant improvement by the addition of elegant floor mosaics and marble veneers on the walls. 6) The seaside prospect of the the villa became magnificent with the construction of an impressive façade. This featured a long pilastered portico (Portico P) flanked by two round projections (Room A at the west, and Room T at the east).

The following sections analyze these architectonic changes, which transformed the villa into a monumental residence for hosting leisure, social, and likely economic activities. It is possible that even religious gatherings occurred in the fourth centuy AD. The floor mosaics so far uncovered will be examined separately in Chapter 5.

83 See Guida, p. 102, figg. 154-155;idem for a detailed analysis of the architectonic changes especially of the phases T7-T9, pp. 45-49; 65-69; 77-81. 30

4.1. The modifications occurred between the third and fourth century AD inside the villa.84

84 After Guida, p.102. 31

4.2. The Western Baths - A- (in red) and the Eastern Bath - B - (in orange).85

85 After Guida, p.38. 32

4.3. The southern area of the Villa at Palazzi di Casignana.86

86 After Guida, p.64. 33

The Western Baths

4.4. The Western Baths modifications (highlighted in red) between the third and fourth century AD.87

Figure 4.4 shows highlighted in red the most significant additions within the Western Baths that took place between the third and fourth century AD. Portico 17 was built using brick columns finished in stucco, on the northern and western side of the villa giving the Western Baths a monumental entrance from the garden. From the portico, entrance into the baths was offered in the following way: a visitor would have access through a door with a limestone threshold into corridor 15-16 (fig.4.2), from which he or she could enter Room 12, the frigidarium. After Room 12, a visitor could move along Corridor 19 toward the calidarium of the western baths, which is composed of Room 24, Room 36 and Room 29. During the fourth century renovation, all the rooms of the Western Baths were more elegantly arranged with the addition of apses and expensive floor mosaics. Also more tubs were inserted in the bath suite. The grand scale and elaboration of the baths suggests that they may not have been for exclusive, private use of the dominus, but were likely available also for his acolytes and clientela.

87 After Guida, p.48. 34

Corridor 15 leads to the frigidarium Room 12. It is paved with a black tesserae mosaic framed by a white band (see fig.4.4.).

4.5. Room 12, frigidarium of the Western Baths.88

Room 12, frigidarium (fig. 4.5). This room was enhanced with four apses at each corner and paved with a geometric mosaic. Tub 21 was added on the eastern side and Tub 22 was added to the western side, where a window offered a fine view of the garden. Room 12 is one of the most opulent and elegant rooms of the entire villa measuring ca. 6 m. x 6 m. while the apses are ca. 2 m. wide. The walls of the room were covered with marble veneers, portions of which are still visible today. At the east and west side of the room there are two tubs (21 and 22),

88 After Guida, p.51. 35

4.6. Tub 21 inside Room 12.89

facing each other, which were used as cold baths. They both have walls covered with marble panels. Tub 21 (fig.4.6) has an oval shape and is smaller than tub 22. The apse of Tub 21 opens with large windows onto Room 9 of the Eastern Baths. Tub 22 has windows opening on the outside garden. In Aquitania, the villas at Sain-Sever, Castelculier, Bourg and Chiragan90 similarly feature two or three tubs in the frigidaria, indicating that these spacious bathing facilities offered to a larger number of bathers the possibility of spending time talking and relaxing, while enjoying immersion in cool water. The central area of Room 12 is paved with an elegant mosaic composed of white, black and green tesserae forming a perspective cube motif.91 At the centre of the frigidarium and near Tub 22 two marble trap-doors indicate the presence of a drainage underneath the floor. Remains of the fallen walls indicate that the ceiling of this frigidarium was domed or cross-vaulted. A great quantity of light blue glass tesserae was found in the room’s debris, suggesting the presence of a blue glass mosaic decorating the vaulted ceiling. The effect must have resembled a blue sky arching above the room, which was enhanced by the sparkling reflection of the water contained in the tubs. Visitors, entering the

89 Photo by author. 90 BALMELLE 2001, p.191 with bibliography. 91 von BOESELAGER 1983: mosaic in Palermo, pl. X.20; mosaics in Solunto pl. XVI.31. 36 room could immediately feel the dramatic effect of the high roofs and windows. The most imposing frigidaria would have had high ceilings, exactly for this purpose.92

4.7. Apse of Room 12 with peacock motif.93

A large number of baths in late Roman houses are characterized by a greater development of the unheated areas in respect to the warm and hot rooms. The cold baths played an important part in the Roman bathing, and were enjoyed after the steam and hot baths. At Palazzi di Casignana, the frigidaria (Room 12 and Room 9/25) are the biggest rooms of the Western and Eastern baths. Inside the fourth century AD villas at São Cucufate (Portugal), and Piazza Armerina (Sicily), the frigidaria are much bigger than the calidaria. In São Cucufate, although there are only the foundations of an unfinished monumental bath, it is possible to determine the intended size of the frigidarium, which was 25 m. long and 12 m. wide. In the villa at Piazza Armerina, the frigidarium has an octagonal plan with six niches, two of which served as doorways. Two provided access to the pools and the other four it has been suggested were

92The octagonal plan of Room 12 has its prototype in the Tower of the Winds, also called horologion (timepiece), which is an octagonal Pentelic marble tower on the Roman in , that perhaps was built by Andronicus of around 50 BC. The octagon recurs inside the Octagon Hall of the Domus Aurea and from then onward is frequently adopted in Roman domestic architecture. See IACOPI, I., “Domus Aurea di Nerone,” in “Functional and Spatial Analysis of Wall Painting,” in “Proceeding of the Fifth International Congress on Ancient Wall

Painting (Amsterdam 8-12 September 1992), Leiden 1993, pp. 59 ff. Besides the frigidarium at Piazza Armerina, other examples of octagonal frigidaria are found at Bulla Regia in (THÈBERT 2003, pp.134-135); at Leptis Magna in Libya (CREMA 1959, p.538), and at Bax Farm, in Kent (Britain). This British villa was excavated in 2006 and uncovered an octagonal room measuring 10 m. across, which contains a pool that is believed could have been used for Christian baptism. See PITTS, M., British Archaeology 91, 2006.

93 Photo by author. 37 probably used as apodyteria for changing.94 The plan and size (10 m. x 10 m.) is very close to that of Frigidarium 12 at Palazzi di Casignana.95 As A. Bouet points out, frigidaria in late Roman villas are typically very spacious rooms because they are cold and refreshing spaces and thus were more suitable for large groups of bathers. People could bathe and relax more comfortably here, and talk or conduct business for longer periods of time.96 One may also hypothesize that other subsidiary activities, not strictly related to bathing, occurred inside the frigidaria as, for instance, eating and drinking. Literary sources describe Roman bathers consuming food and drinks inside the public baths.97 The same may have occurred inside private baths. Yet, studies on Roman bathing tend to not consider the movement of the people inside these facilities, and often do not further explore where and how different activities may have occurred. The four apses of Room 12 are clearly marked off from the central floor by a geometric motif of -meander alternated with black squares. Their field is decorated with a black and white peacock tail motif (fig.4.6). Each apse being sufficiently wide (2 m.) could accomodate a portable stibadium for two people placed against the wall.98 It is conceivable that inside a domestic frigidarium bathers gathered also to eat or drink using portable tables, and all kinds of cookery that servants provided when needed. At Armerina, a frigidarium similar in plan to Room 12, also features apses that are graphically separated from the rest of the room by mosaic strips. Supposedly, they were used for changing clothes.99 The architectural remains invite us to interpret the choice for particular shapes and plans in relation to space function. Recent studies on Roman domestic architecture, such as that by P. M. Allison, have paid closer attention to the fluidity of room function that was made possible also by the easily movable Roman furniture.100 By looking at the architecture alone, and especially when artefacts emerge, it is possible to gain a more nuanced understanding of what activities occurred inside Roman baths, both public and private, since the practice of bathing for the Romans meant more than just cleansing the body. It was a very important aspect of private life but also a custom that had a strong social component. While bathing, important negotiations were conducted or influential friendships were made. The literary sources mention various

94 CARANDINI, A.- RICCI, A. - DE VOS, M., Filosofiana. La Villa del Casale di Piazza Armerina. Immagine di un aristocratico romano al tempo di Costantino, Palermo 1982, 326-73; BAUM-VOM FELDE, P.C. Die geometrischen Mosaiken der Villa bei Piazza Armerina, Amburgo 2003; WILSON, R. J. A.: Piazza Armerina, Granada Verlag, Londra 1983; SETTIS, S., "Per l'interpretazione di Piazza Armerina", in Mélanges de l'Ecole Française de Rome. Antiquité 87, 1975, 2, pp. 873-994.; PACE 1951 and 1955. 95 WILSON 1983, pp.21 ff. 96 BOUET 1997-1998; idem, 2004. ALARCAO, J., ÉTIENNE, R., MAYET, F., Les villas romaines de São Cucufate (Portugal), Paris, 1990, pp. 114-116. 97 For the consumption of food see Martial, Epig. 12.19; for drunken behaviours see Martial, Epig. 12.70; Seneca, Ep. 122.6, Pliny, NH, 14.139, and Quintilianus, Inst. 1.6. 44. 98 In Rome, inside the catacomb of Peter and Marcellinus, the arcosolium (a burial niche surmounted by an arch) of cubiculum 76 represents a boy and a man reclining together at the stibadium. See DUNBABIN 2003, p.180, fig.105. This painting suggests that it was possible that smaller spaces like the apses of Room 12 were occupied by two diners. 99 WILSON 1983, pp.21-22. 100 ALLISON, P.M., Pompeian Households: An Analysis of the Material Culture, University of California, 2004; idem, “How Do We Identify the Use of Space in Roman Housing,” in Functional and Spatial Analysis of Wall Painting, in E.M. Moorman (ed.), Leiden: Babesh, 1993, pp.1-8; idem, “Artefact Distribution and Spatial Function in Pompeian Houses,” in The Roman Family in Italy: Status, Sentiment, Space, in B. Rawson and P. Weaver (ed.), Oxford: Clarendon, 1997, pp.321-354. 38 activities that occurred within the public baths.101 The architecture and decorations of private baths had to be visually attractive but, more importantly, they had to respond to a variety of uses and social interactions.

4.8.Corridor 19. Mosaic.102

Corridor 19 (fig.4.8). This is the second corridor symmetrical to corridor 15 located at the south side of the frigidarium (Room 12). The walls of corridor 19 are covered with marble veneers. The floor is paved with polychrome geometric design in opus tesselatum, featuring a peltae motif.103 At the centre of the floor, a vegetal motif is inserted in geometric figures. From corridor 19, one progresses to the heated part of the Western Baths (calidarium). But from here one can also enter Room 20, which connects the Western baths with the Eastern Baths.

101 CARCOPINO 1940 refers to the following texts: Libanius, Or. 3.6.16 on rooms for lectures, poetry reading and music. Numerous sources discuss eating in the baths; Martial, Epig. 5.70 and 12.19; Seneca., Ep. 56.2. Drinking: Martial, Epig. 12.70; Pliny, NH, 14.139. Prostitution: Martial, Epig. 3.93.14; , Hipp. 5. 102 After Guida, p.50, fig.53. 103 Guida, p. 50, fig.53. 39

4.9. Room 24.104

Room 24 (fig. 4.9). At the westernmost end of Corridor 19 there is a threshold into Room 24, which together with room 36 and 29 form the suite of heated rooms (calidarium), of the Western Baths. Room 24 is rectangular and features two apses on the short sides that were occupied by benches made of stone or marble. The floor was built on suspensurae and is decorated with a sophisticated geometric tesselatum. The mosaic features a white background where a motif of black squares stands out along crosses and elongated hexagons containing various geometric figures. Remains of many glassy tesserae found scattered in the area suggest the presence of a mosaic on the ceiling similar to Room 12 of the frigidarium. All three rooms that form the calidarium feature visible tubuli running up the walls to distribute heat. During the fourth century AD, they originally were hidden by marble slabs, parts of which remain in place. The thresholds of each room are cut diagonally in order to prevent the dispersal of the heat. This is a common architectural expedient in warm baths.

104 Photo by author. 40

4.10. Room 36, calidarium.105

Room 36 (fig. 4.10). This room is in the suite of heated rooms, adjacent to Room 24. It has a square plan and features two tubs (32 and 33). These were heated, as indicated by the presence of tubuli, suspensurae, and praefurnia placed behind them. Tub 32 is located on the western side of the room and has a rectangular shape with two curvilinear sides. Tub 33 is located on the southern side and is also rectangular. The sides of both tubs are veneered with marble slabs to match the walls of Room 36. The floor of Room 36 rests on suspensurae and is paved with a complex, polychrome geometric mosaic.

105 Photo by author. 41

4.11. Room 29 (calidarium).106

Room 29 (fig. 4.11). On the eastern side of Room 36 there is a threshold into Room 29, the third room of the calidarium. The door jams are cut diagonally to prevent dispersion of the heat. The floor of the room is carried by suspensurae and tubuli run up inside the walls. The room has an octagonal shape and is paved with a opus signinum that covers a damaged mosaic floor featuring a polychrome motif of entwined circles containing stylized flowers. The ceiling was likely decorated with a glittering mosaic since many glassy tesserae were found in the debris of this room.

106 After Guida, p.55. 42

4.12. Praefurnia (furnaces) of the Western baths: Rooma 30 and 35.107

The Praefurnia of the Western Baths (fig.4.12). The furnaces that heated the calidarium of the Western Baths are located at the south-western corner of the villa. The entrance to the praefurnia is from an open space at the south side of the Western Baths. A set of three steps leads to a first rectangular room (Room 30, fig.4.12) that is built partly underground. At the northwest corner of Room 30 are located the foundations of the heater that warmed Room 29. On the western side of Room 30 lies a vaulted and curved gallery of furnaces that provided heat for the other rooms of the calidarium, namely Room 24, and Room 36 with its two Tubs 32 and 33. The heaters for these rooms were placed on simple brick bases. At the end of Gallery 35, in correspondence with Room 24, the vault features an opening for the passage of air. The floor of the gallery was made of sandy lime and stone slabs. Along the gallery runs a drain conduit made with pitched tiles.

107 After Guida, p.38. 43

The rooms that connect the Western Baths with the Eastern Baths

4.13. Room 20.108

Room 20 (fig.4.13). From the Western Baths, one could enter the Eastern baths by taking corridor 19 and crossing Room 20 (fig. 4.12). This is paved with a mosaic featuring a swastika and square motif in black and white tesserae.109 Tubuli run up at each corner of the room, suggesting a function as tepidarium.

Room 18 (fig.4.12) is a remaining space of an earlier plan of the Western Baths linked to Room 20 through a door placed on the east side wall. In the fourth century AD, with the latest arrangement of the baths, Room 18 appeared disconnected and in an isolated position from the rest of the Western Baths. During this time, the marble slabs that originally covered its walls were removed and perhaps reused for the restoration of other areas of the villa.

108 Photo by author. 109 Guida, p. 53, fig. 58. 44

The Eastern Baths in the 4th c. AD.

4.14. The Eastern Baths (in orange).110

110 After Guida, p.38. 45

4.15. The Eastern Baths: the new additions (highlighted in red) that took place between the 3rd and 4th c. AD.111

The Eastern Baths (or Area B in fig.4.14) were also enlarged and elegantly decorated during the latest renovation of the villa. As the Western Baths of the villa, they feature a frigidarium and a calidarium composed of the following rooms in sequence: Corridor 13; Room 9 and Room 25, which formed the frigidarium. Room 10, 11, and 37, which formed the calidarium.

Corridor 13 (fig.4.14). The entrance to the Eastern Baths is through Corridor 13. This runs adjacent to corridor 15 of the Western Baths and to Portico 17. Corridor 13 was paved with squared bricks.

111 After Guida, p. 46. 46

4. 16. Room 9: Nereid on mosaic floor.112

Room 9 (fig.4.15).113 It is the frigidarium of the Eastern Baths. It features a rectangular plan measuring 9 m. x 7 m. This room is paved with an opus tesselatum representing a thiasos of four , each respectively riding a lion, a bull, a horse and a tiger. The four beasts have a tail ending with three fins, characteristic of sea monsters. The mosaic is made of large tesserae of green and white colour with different shades. It was damaged, and never repaired, when a drain conduit was installed during the latest phases of the villa baths.114 On the northern side of the room, there is a large rectangular tub, Tub 8, (see plan on fig.4.15) with a parapet and two steps. On the back wall of the tub there is a hole in which the water pipe was inserted. Also the mark left by the water (1.10 m. high) is evident all around the walls of Tub 8. At the west side of the tub there is a little space with a hole where a bench was placed. Perhaps this part was occupied by a latrina.

112 Photo by author. 113 See Chapter 5 for a description of this mosaic. 114 Guida, pp.56-57. 47

4.17. Room 25 with opus sectile floor.115

4.17a. Room 25: Tub 26 capable of holding 20, 600 liters of water.116

115 Photo by author. 48

Room 25 (fig. 4.17-4.17a). It is accessible from Room 9 through a wide entrance. This room is paved with an exquisite marble opus sectile117 and features walls covered with marble veneers, portions of which are still in place. The upper parts of the walls were originally decorated with frescos as indicated by various fragments of red, yellow, pink and orange plaster found in the debris. The lavishness of the decoration extends to Tub 26 (fig.4.17a), which was used for cold baths, and located on the south side of the room. A terracotta pipe underneath the tub floor runs from the eastern wall of the tub and goes deep underground to the sea for drainage. A series of little wells that regulated the water drain have been uncovered along the drain-conduit. Tub 26 (fig.4.17a) is huge and has a round shape; it could contain 20.600 litres of water. It is paved with a monochrome mosaic that is still in place. Its walls were originally covered with 1.50 m. high marble slabs, portions of which still stand in place. Sidonius Apollinaris (ca. 430-489) in one of his letters when recalling the frigidarium of his villa at Avitacum in Aquitania, described the pool as worthy of a public bath.118 Tub 26 at Palazzi di Casignana with its huge dimensions falls in the same category illustrated by Sidonius. On the south side of the tub there are three window seals indicating the presence of three windows opening onto the garden, the view of which seems to have been important to appreciate while bathing.119 Several glassy polychrome tesserae, mainly of blue colour, were found in the debris suggesting a mosaic decoration of the vaulted ceiling in imitation of the sky even inside this room. Moreover, the marble rhomboidal and rectangular tesserae that have been recovered from the rubble of this room may have belonged to an inlaid decoration of the walls’ upper parts. At the southwest corner there is a small Tub 27 located in a small room that was originally painted in red and pink as uncovered fragments of painted plaster indicate. The drain pipe of the tub was placed in the fourth century AD across the southwest corner of Room 25 causing substantial damage of the opus sectile floor, which was consequently repaired with recycled marble. The large dimension of Room 25 with its monumental and lavish elements (an imposing tub, a rich opus sectile floor and a likely impressive ceiling), certainly makes the wealth of the villa owner visible to his associates, enhancing his prestige and power. From the outside, the structure must have had a striking appearance. Late Roman baths were typically roofed with domes. V. de Nittis reconstructed the roofs of the baths at Palazzi di Casignana as represented on fig.4.18, and this seems a very plausible reconstruction. The African polychrome mosaic of the landowner dominus Julius found at , Tunisia (fig. 4.19), which is dated to the late fourth century AD, also depicts a late Roman villa roofed with domes from which smoke comes out indicating the presence of baths.120

116 Photo by author. 117 For a description of this floor see Chapter 5 of this study. 118 Sidonius, Epistulae II, 2, 5. 119 See Guida, pp.56-57. 120 Guida, p. 111. BLANCHARD-LEMÉE et Alii, Mosaics of Roman Africa: Floor Mosaics from Tunisia, 1996. 49

4.18. Reconstruction of the Baths’ roofs by V. de Nittis.121

121 After Guida, p. 110, figg. 166-168. 50

4.19. Tunis, Musée du Bardo. Mosaic of Dominus Julius (late 4th c. AD).122

Room 10-11 and 37 (fig. 4.15). These rooms form the heated suite of the Eastern Baths. Room 10 has a rectangular plan measuring 3 m. x 5 m. There is an apse on the southern side. Terracotta tubuli run up inside the walls and the floor is paved with recycled marble tiles. Room 11 is adjacent to Room 10. It also has a rectangular shape and presents similar characteristics to Room 10: an apse on the southern side, tubuli inside the walls and a floor paved with recycled marble tiles. Room 37 is the third room of the calidarium. It features a square plan measuring circa 3 m. x 3.5 m. The room is paved with marble recycled tiles and it is flanked by two semicircular tubs, 38 and 53 (fig. 4.15), which were heated by tubuli.

The Praefurnia of the Eastern Baths: Rooms 23 and 39 (fig. 4.12). Two are the furnaces for the Eastern Baths. The first (23) is placed partly underground at the back of Room 10, 11, and 37, which form the calidarium. Its entrance is at the south through a pilaster portico. Inside Praefurnium 23 there are two facing pairs of half pilasters that were probably part of the roof support. The second praefurnium (Room 39) is located at east of tub 38 and is composed of two communicating spaces.

122 After Guida, p.111, fig. 169. See DUNBABIN 2003, p.120, fig. 122. 51

4.20. Six unadorned rooms (in dark blue) of the Eastern Baths.123

Rooms 1, 3, 4, 5, 51, and 54 (fig.4.20. ) At the northern part of the Courtyard 55 a group of six rooms are paved with plain opus signinum and feature no elegant decorative or structural elements. Rooms 2 and 6 are not floored at all. Because of these characteristics, the archaeologists have hypothesised that all of these rooms were used for service to the baths facility. Portico 17 runs at the northern and western side of these rooms. Portico 52 lies at the eastern side connecting this area with Courtyard 55. The fact that these six rooms are framed by two porticoes, which were typically used for exercising, suggests that they were closely related to the practice of bathing.

123 After Guida, p.38. 52

4.21. Room 7, perhaps a dining room.124

Room 7. A remarkable embellishment at this time took place in Room 7 that gave precedence to this space within the whole bath quarter of the villa. 125 The western entrance that put Room 7 in communication with Room 9 was walled, whereas the eastern entrance was enhanced with a brick column coated with stucco. The marble basis of this column was found in the debris of the room. In front of the entrance a portico with pilasters (52) was created by tearing down a wall that originally framed the eastern side, and by leaving three equidistant pilasters made of bricks coated with stucco. A flight of four steps was also added on the eastern side of Portico 52 that created a monumental access to Room 7 from the peristyle, Courtyard 55. The floor of Room 7 was paved with an elegant polychrome geometric mosaic featuring a four star motif. It is very likely that this room functioned as a triclinium,126 or dining room, since it is located in a particularly privileged position near Courtyard 55 and provided with a view of the garden on the east side. The column placed at the room’s entrance underlined the importance of this space. Likewise, the refined and complex polychrome floor mosaic and the lost painted walls strongly point to the representative role played by Room 7 within the thermal facility of the villa. The placement of an elegant dining room in the bath quarters of the villa would be appropriate since

124 Photo by author. 125 Fase T7, see Guida, pp. 67-68. 126 The name triclinium comes from the use of three sloping couches, the klinai, ranged around the sides of the room while a squared table stood in the middle. Diners would recline on these surfaces in a semi-recumbent position. The fourth side of the table was left free, presumably to allow service to the table. The Romans considered the reclining position more comfortable and also a mark of elegance and social distinction whereas the lower classes and slaves would dine seated.. 53 the Roman custom was to dine after exercising and bathing.127 At times triclinia are recognized by plain white panels on the mosaic floor of the room on which the dining couches would have been placed. The floor mosaic of Room 7 does not feature any panel for couches and is not laid out in a T or U shape. Yet, its architectural and decorative features point to a reception use of this room. The term ‘reception room’ is commonly used to designate one or more rooms of a house that are distinguished from others for their size, architectural development and decoration.128 These rooms are essential components of wealthy residences. The elite had the obligation to host guests and display the high social status through the lavishness of their home decoration and architectural grandeur. Dinners played a central role in Roman life. Entertaining guests was a matter not only to meet friends but more importantly to create political ties and exhibit social status. A dominus would often invite his clients who in return gave him political support.129 In most cases the reception room in late Roman houses is easily identifiable by several elements including its location on a major axis of the house, the proximity to a main entrance or peristyle, and the elegant decoration. The orientation of reception rooms does not always follow a rigid rule, regardless of Vitruvius’ instructions to place them in an area of the house that is exposed to a good climatic condition.130 From archaeological evidence more importance seems to have been paid to their location in proximity to the peristyle of the house, and through it, to the house’s main entrance. When a reception room is located on a lateral wing of the house, it is often because it was added to the original nucleus of the house in a later phase of construction.131 In late Roman houses, especially from the late third century onward, dining rooms were provided with an apse for the setting of the semicircular couch called stibadium, which came to replace the traditional three rectangular couches.132 In large houses especially of the fourth century AD spacious and elegantly decorated rooms frequently served a double function, that of a triclinium (dining room) and that of a reception space. At the villa of Piazza Armerina and at the Villa at Patti Marina, these two functions appear separated in two or more rooms. However, in most cases the reception room is just one and stands out among other rooms of the same complex that are smaller, more simply decorated or totally unadorned. Its larger dimensions, elegant decoration of floors and walls, and location in a privileged position of the house make it easily recognizable to the visitor. The great diversity of reception rooms in the late imperial domestic architecture limits an exact identification of their function, as it is the case of Room 7 at Palazzi di Casignana. Moreover, in Roman times within domestic architecture some rooms served multiple functions. The Younger Pliny when describing his villa in Tuscany, mentions a room that functioned as a dining room but also a reception room for his friends.133 The same polyvalence occurs in

127 YEGÜL 1992. 128 For a discussion on reception rooms see BALMELLE 2001, pp.155-157; THÉBERT 1985, p. 350; BALDINI LIPPOLIS 2001, p. 57. 129 See D’ARMS, J., “Control, Companionship, and Clientela: Some Social Functions of the Roman Communal Meal,” Studies in Roman Society, (EchCl 28, n.s. 3, 1984), pp.327-348. 130 Vitruvius, De Architectura, VI, 4. 131See for instance reception rooms in two houses of Djemila: the Maison D’Europe and the Maison de , in BALDINI LIPPOLIS 2001, La Domus Tardo Antica Maison d’Europe, p.190; eadem, Maison de Bacchus, p.193. 132 For more details on the use of stibadium see Chapter 5 of this study. 133 Epist. 5.6.21, II, p.66: in hac diaeta...cotidiana amicorumque cenatio. 54 domestic spaces of fourth century AD houses. As Balmelle points out,134 Sidonius Apollinaris mentions a number of villas’ reception rooms without knowing exactly when to use one term or the other. For example, in Epist. 2.2.11, II, p.49 he writes: diaetam sive cenatiunculam. In other passages, Sidonius describes a number of rooms that were used for receiving guests: alta cenatio (high dining room);135 hiemale triclinium (winter dining room) 136 deuersorium (living room or rather guest apartment). 137 It is difficult to apply each of these terms to a specific architecturalspace inside any given villa. So far, inside Villa at Palazzi di Casignana, beside Room 7, three other reception rooms have been identified, all three in the seaside area of the villa: the apsidal cruciform Room I with a polychrome floor mosaic; Room B that features the Four Season mosaic; and Room Q paved with the mosaic. Their recognition is mainly due to their floor decoration, marble wall revetments and prominent location in the whole villa plan. It is not possible to say when and how they were used but they clearly stand out within the general architecture of the villa.

134 BALMELLE 2001, p. 156. Sidonius, Epist., 2,2,11, II; See commentary by REBUFFAT 1970, p. 318. This scholar underlines that the term diaeta does not have the classical meaning but here refers to a private dining room. 135 Sidonius, Carmen, 22.207. 136 Balmelle 2001, p. 156. 137 Sidonius, Epist., 2.2.13. 55

The changes in the southern area of the villa between the third and fourth century AD.

4.22. The southern area of the villa (in yellow).138

138 After Guida, p.64. 56

4.23. The 4th c. additions (highlighted in red) to the southern area of the villa.139

Portico 59 (fig.4.23). The southern area of the villa was enhanced through the construction of Portico 59 supported by pilasters, which opened onto the main peristyle of the villa (Courtyard 55). It was decorated with a geometric polychrome mosaic of small tesserae. Future excavations may clarify what was the total length of this portico towards the eastern residential area of the villa.

Room 58 and 60 (fig.4.23). In an earlier phase of the villa these two rooms formed one space. When the fourth century renovation took place they were divided by a wall into two separate rooms. Room 58 was used as a passage room or standing place between Latrina 40 and Portico 59. Room 60 has preserved its floor mosaic with a small black tesserae on a white background

139 After Guida, p.68. 57 that form different geometric motifs interrupted by marble slabs that are not longer in situ but which left the imprint on the floor plaster.

Some observations on Portico 59

The construction of walkways or porticoes, as Portico 59, in Roman villas was dictated by a diffused taste of blending nature with elegant architecture. While offering a shelter from the sun’s rays or rain it was important to be able enjoying fresh air and natural views of gardens and the outdoor. The porticoes of the villa help to organize the space and provide the villa with an elegant front. Porticoes, as well as corridors, are elements of transition between one area and the other of a residence. Their decoration played an important role at underlining the hierarchical role of the rooms inside the house. The more elaborately decorated a corridor or a portico was, the closer one knew to be to a prestigious area of the house. Baldini Lippolis cites several fourth century AD residences outside of Rome that feature a long portico, or xistus, that connected several pavilions of a villa as Portico 59. These include the villa at Piazza Armerina, the Domus Faustae, and the Villa of Maxentius.140 Some may have been used for sport training while they could also have constituted a sort of ceremonial way for clients and guests of the owner towards monumental spaces such as the baths, apsidal rooms, and triclinia. Porticoes framed courtyards of Roman villas in the manner of gymnasia and palaestrae recreating a place for literary pursuits and cultural refinement. The Greek gymnasium was a facility where athletes trained to compete in public games but it was also a place for socializing and engaging in intellectual activities. Cicero called Academia his villa with the gymnasium in reminiscence of the Platonic school of the Athenian Academy.141 In the dialogue of the De Oratore, which is placed at the Villa of Licinius Crassus in Tusculum, Cicero explains that Licinius’ villa is a well suited place for cultural conversations since it has porticoes, palaestrae and benches that recall the Greek gymnasia and the discussions that happened in them.142 Many villa owners had libraries in their villas that underline the role of villas as centres for intellectual leisure. Horace writes that the gymnasium and the palaestra created the right atmosphere for literary activity.143 In addition to floor mosaics, frescoes and wall marble veneers, it was highly likely that Villa at Palazzi di Casignana was decorated with marble statues, all kinds of movable furniture and artworks. Their absence in the current archaeological record may be due more to pillage, destruction, and recycle of building material such as marble than to a real lack of them. In fact, it is frequently attested in many imperial villas by both the archaeological evidence144 and ancient sources145 that the very wealthy Romans took particular care at furnishing with sculptures and artworks their villa’s porticoes, gardens and also reception rooms. These areas became places of self-representation of the villa’s owner. Cicero wrote to his friend Atticus who was in Athens at that time to purchase on his behalf decorations for his gymnasium and lamented his disappointment with somebody who had purchased on his behalf statues of Mars and Bacchus

140 BALDINI LIPPOLIS 2001, p.68; DE ANGELIS BERTOLOTTI et Alii 1988. 141 Cic. Tusc.2.9. 142 Cic. De Orat. 2.9.10. 143 Hor. Sat.6.72 ff. 144 At the Villa in Piazza Armerina a few fragments of statues survive. See WILSON 1983, p. 33. 145 For studies on display of statues and decorative plans in villas see BARTMAN 1988; NEUDECKER 1988; BECATTI, G., Arte e gusto negli scrittori latini, Firenze 1951; STIRLING 1996. 58 which did not fit the context of his palaestra (Att.1.8.2). Pliny wrote in one of his letters about the eques Terentius Iunior: ‘One would think that the man lives in Athens, not in a villa.’146 Even in Late Antiquity sculptures continued to be important components of the domestic décor. Several statue collections are known from fourth and fifth century houses. Statues of Meleager and , for instance, have been found at in Syria, in Southern France, 147 and at in .148 Three statues of were found in three houses from the same areas, inside a triclinium at Ostia, in Southern France and Ptolemais in Cyrenaica.149 The most popular divinity was Venus represented in seven statues from four houses (Antioch, Ostia, Ptolemais, and Southern France).150 A statue of Jupiter and Ganymede was found inside or near a triclinium in the house of the Greek Charioteers in Carthage.151 Unfortunately, only in a few cases has it been possible to determine where they originally stood. At the Villa in Piazza Armerina, bases of statues were found in the triconch and in the enclosure wall of the adjoining oval peristyle.152 Statues of gods and mythological heroes reflected the ideals and the power of the dominus, thus contributed to his glorification. At the Villa of Desenzano in Northern Italy several statues of Hellenistic subjects have been found inside an underground passage below Peristyle 2. Some had been damaged in antiquity and others were incomplete. It has been suggested that during a possible Christian ownership of the villa they were thrown in the underground passage because not longer valuable.153 An aspect that has not been much explored is the possible diffusion inside late Roman houses of sculpture representing Christian themes. There is evidence of a Christian sculpture production parallel to the pagan- classical. Small statues with Old and New Testament subjects have been found although their original context is not certain.154 At Palazzi di Casignana, the villa’s architecture with its monumental structures and articulated plan reveal the intention of the dominus to boast power and prestige.155 It is highly likely that this message of dominance that the architecture conveys to the viewer was also complemented with sculpture and all kinds of works representing themes and subjects that exalted the villa’s dominus’s virtues and pursuits. If the owner of the villa was converted to Christianity, it is not excluded that his faith was also reflected in the decorative apparatus of the villa that is not longer extant. Porticoes, such as Portico 59, being frequented areas of transition within the villa’s plan were among the most suitable parts where artwork’s display, whether pagan or Christian in nature, could produce a very effective result.

146 Pliny, Ep. 7.25.4. 147 ELLIS 1991, pp. 127-128, with note 4, 55, 56, 57 and 58. For the statues of Southern France and Antioch see BRINKERDOFF, D., A Collection of Sculpture in Classical and Early Christian Antioch (New York 1970). 148 See PESCE ,G., “Il Palazzo delle Colonne” in Ptolemaide di Cirenaica (Monografia di archeologia libica 2, Rome 1940), pp. 80-89. 149 ELLIS 1991, p.133, note 57. 150 BARTMAN 1991, pp. 71-88, especially citations in nn.11-14; 31-32; and 67. 151 GAZDA E.K., “A Marble Group of Ganymede and the Eagle from the Age of Augustine,” in HUMPHREY J.H. eds., Excavations at Carthage, 1977, Conducted by the University of Michigan 6 (Ann Arbor 1981), pp.150-166. 152 WILSON 1983, p.33. 153 SCAGLIARINI CORLAITA et alii, Villa Romana di Desenzano,1992, pp.100 ff. 154 Baldini Lippolis refers to the interesting collection of 11 small statues kept in the Metropolitan Museum of New York that perhaps came from Asia Minor. See LIPPOLIS 2001, p. 89, note 458. See also Age of Spirituality 1979, pp.406-411 and nn. 362-368. 155 On message of power reflected in late Roman villas see ELLIS 1991. 59

Latrina 40 (figg. 4.23 and 4.24). This latrina is located in the south-western side of the villa within a group of rooms that perhaps had a service function. It features a round plan and was entered from a rectangular room (41) located on the eastern side that functioned as a vestibule where a few people could stand. The floor was paved with white marble slabs, portions of which are still preserved in the centre of the room. On the perimeter wall (about 7 m. long) is visible the bench on which the seats were placed side by side, positioned at 70 cm. distance from each other. In front of them runs a water channel that was used to wash the sponges (the Roman equivalent of toilet paper). Seven seat dividers are still visible. It is likely that Latrina 40 could accommodate about 10 people. A small lavatory lies on the north-western corner of the room which was furnished by water pipes. In an earlier phase room 41 was connected to an open space that led to a second smaller latrina (42) located on the opposite side of Latrina 40, which may have been used by women.156 The only remains of this second latrine are a seat bench and the drainage canal. This room was eliminated during the latest phase of the villa (Fase T9),157 when the open space that connected the two latrines was roofed and transformed into a narrow corridor. Noteworthy is the sophisticated water and sewage system of the Villa at Palazzi di Casignana. A series of wells used to inspect the water drainage has emerged around the villa especially near the baths and latrines. 158

156 The same has been suggested in regards to two private latrines of different size at Herculaneum: HOBSON 2009, pp.81-82. For gender taboos about using the toilet within a Roman household see NIELSEN 1990. 157 Guida, p.69. 158 Guida, p.71.

60

4.24. Latrina 40 with marble pavement.159

At Palazzi di Casignana the latrina 40 provides a distinctive example of lavish latrine where very expensive marble was used to pave the floor. In Roman upper class houses latrines were not considered secondary rooms. Latrines had a socializing role in Roman culture.160 An epigram by Martial (II. 77) is revealing about what occurred inside a Roman public latrina. “Vacerra dallies for hours, and sits a whole day in all the [public] latrines. Vacerra wishes to dine, not to empty his bowels.” This passage shows that Romans considered latrines not only sanitary places but also meeting points where people may have spent time indulging, as Vacerra did, in hope of receiving an invitation to dinner. Latrines and what occurred in them have not been subjects for archaeologists and researchers of Roman architecture. Yet, literary sources recount that public latrines were places were senators and business men met and chatted about ordinary and important activities.161 In order to understand the importance of late Roman latrines in domestic architecture it is necessary to turn to the public sphere that much influenced the private.

159 Photo by author. 160 For research on private toilets in Roman houses see JANSEN, G.C.M. “Water Systems and Sanitation in the Houses of Herculaneum,” Mitteilungen des Leichtweiß-Instituts für Wasserbau der Technischen Universität Braunschweig 117 (1992), pp. 449-68. For a general treatment of sanitation in Roman Italy, see KOLOSKI- OSTROW, A.O., The Archaeology of Sanitation in Roman Italy: Water, Sewers, and Latrines, (Chapel Hill, forthcoming). 161 There is a vast literature on the social aspects of Roman bathing. See for instance, FAGAN 1999. YEGÜL, F. Bathing in the Roman World, 2010 (forthcoming). 61

There is evidence that the Romans built communal multi-seats latrines from at least the second century BC which could accommodate even 100 people.

Latrine seat (after CONNOLLY, P. y DODGE, H., La Ciudad Antigua. La vida en la Atenas y Roma clásicas, Madrid, 1998). Along the walls a steady marble bench extended with a series of holes with a narrower drop-shaped opening in the front; these holes were distant enough to leave things between users. On the ground the water flowed through small channels with inclination enough to allow the water steadily flowed; near these channel there were some buckets with brushes; these brushes had a Hypothetical reconstruction of latrines in Rome, wooden handle and a ball of bath sponge that maybe to 100 people, in the present Largo the Romans used like our modern toilet paper; Argentina (after CONNOLLY, P. y DODGE, after using the brushes, they cleaned them in H., La Ciudad Antigua. La vida en la Atenas y the water channel. Roma clásicas, Madrid, 1998).

62

Neudecker’ remarkable study, Die Pracht der Latrine, describes several well-preserved samples of latrines from Rome and other cities of the Empire and traces the development of Roman latrine architecture.162 This scholar argues that luxurious latrines both in the public and private domain appear in the second century AD. Although the basic technical features such as seat-size, the distance between seats, plumbing arrangements, and so on, remained unchanged from the latrines of the Republic and early Empire, the splendour of their interiors increased dramatically from the second century onward. More care was placed towards ventilation, light, water, and attractive form. Neudecker’s study maintains that within the domestic sphere wealthy owners of the Republic and early Empire preferred portable chamber pots.163 Only from the second century onward, contemporary to what occurred in the public latrine, well off patrons provided elegant bathrooms in their lavish residences for their family and also for their guests. Thus, domestic latrines were not only sanitary facilities but also gathering places for relaxing and networking. Neudecker contends that the luxury displayed in latrines from the second century onward reflects a deeper concern of the Romans for sanitation and hygienic conditions. The diffusion of Roman medical literature forced a change in behavioural norms especially among the aristocrats. Medical writers and philosopher-doctors, such as Galen (ca. 129 - 216 A.D.), Celsus (end of the second century), Epictetus (ca 55–135 AD), generated a deeper awareness among the elite about medical cures and physical care greatly influencing the lifestyle and daily routines in the elevated circles of Roman society. These medical writers encouraged their aristocratic patrons to take care of their physical health in a manner that became almost pathological.164 The aristocratic male had to include in the ethos of nobilitas also the care of physical functions in order to have a better command of his whole personality. Latrine 40 of the Villa at Palazzi di Casignana certainly falls in the category of luxury latrines and bears witness of important cultural changes among the Roman elite of the late Empire.

162 NEUDECKER, R., Die Pracht der Latrine. Zum Wandel öffentlicher Bedürfnisanstalten in der kaiserzeitlichen Stadt. Studien zur antiken Stadt, 1994. Neudecker points out that typically public latrines were found in beautiful locations among the most frequented places in the city. This is contrary to what occurs in contemporary time. Public latrines were integrated into the commercial and social centres of Roman cities and expensively embellished with relief sculptures, statuary, mosaics, and frescoes. A great amount of money was spent by the emperors in order to make the public latrines comfortable and beautiful. Neudecker explains the effort at achieving comfort by building large toilets with wide and spread seats; the elegance and luxury was obtained through sculpted marble armrests, wall paintings, floor mosaics and all kinds of artworks. Neudecker examines the cities of Vesuvius and points out that although Pompeii preserves the best architectural evidence for early imperial hygienic standards the city does not feature luxury latrines. These well appointed facilities flourish later, between the second and third century A.D., and then fade. Under the influence of Christian morality, especially from the fourth century onward, the public latrine reverted back to a mere utilitarian facility, gradually losing both its luxurious form and social role in civic life. A more recent publication on Roman latrines is by HOBSON, B., Latrinae et foricae: toilets in the Roman world, 2009. 163 NEUDECKER 1994, p. 17. 164 Ibidem, pp. 33-34. 63

The route taken by the bathers inside the fourth century Villa at Palazzi di Casignana

4.25. The route of the bathers reconstructed by the villa’s archaeologists.165

165 After Guida, p. 108, fig.163. 64

4.26. The Baths of the Villa at Palazzi di Casignana.

The route taken by bathers has been reconstructed by the villa’s archaeologists (fig.4.25).166 It started from Portico 17 (fig. 4.26), which was the entrance to the baths. From here bathers would move to Corridor 15, then Room 12 (frigidarium), enter Corridor 19 to finally enter Room 20, which was likely the changing room (apodyterium) for both baths being located between the two suites. The same function may have been served by the adjacent Room B, or Room 18. Afterwards, bathers would move to the tepidarium T, calidarium C, the laconicum L, or a sweating room, and lastly to the frigidarium F (fig.4.25). To use the baths in the correct sequence (T-C-L-F), the bather had to cross the same rooms where they had already bathed.167 The baths at Palazzi di Casignana follow the standardize architectural scheme found in both private and public baths all over the Empire, which is defined by Krencher and Kröger ‘der Reihntyp.’168 There was a changing room, or apodyterium, from which the bathers entered the

166 Ibidem. 167 Ibidem. 168 KRENCHER-KRÖGER 1929, p.177, figg. 350, 353, and 354. This traditional method of bathing is reported in the following sources: Pliny, NH 28.55; , Sat. 28; and Mart., Epig. 5.42. Medical authors, such as Celsus and Galen (second century AD), suggest different salutary hydrotherapies that do not follow the ‘traditional’ 65 warm room, or tepidarium. Then, they would move to a hot room, calidarium, and end in the cold room, the frigidarium. The warm rooms were heated by the hypocaust system and around the edge of the floor clay pipes (tubuli) conducted the hot air up the walls. In many houses, and especially in villas, private baths were located at the front of the building, underlining the priority of bathing in the sequence of actions that occurred during a social gathering hosted by the domini. First the owners and their friends would bathe, relax, talk and conduct businesses, and then everyone would move to the reception area for the afternoon meal.169 At Palazzi di Casignana, Room 7 was likely the reception-dining room where the intimate friends of the dominus would move after bathing. The separation of the baths from the seafront area of the Villa at Palazzi di Casignana -H may have been created to reduce the risk of fire, which was alimented by the praefurnia. The baths placed at a distance from the living areas are found for instance at the Sicilian villas of Piazza Armerina and Patti, at Porto Saturo in Puglia, and at Desenzano in northern Italy.170 Both suites of the baths at Palazzi di Casignana stand right behind Portico 17. At the villa in Piazza Armerina171 and at the villa at Montmaurin172 the baths are also located behind the entrances formed by a semicircular portico. The travellers or the visitors who arrived at these villas would have entered the baths and enjoyed some relaxation before dining. It is possible that Area G (fig. 4.27) that included a building with a portico was used for undressing, exercising and gathering before the baths. Games and exercising preceded and reinforced the salutary effect of the baths. They were admitted as long as they were not practiced as spectacle, but for the health of the body and fitness.173 The growing influence of Christianity had an impact also on the practice of exercising. As F. Yegul states, “none of the baths built after the fourth century AD combined a bath and a palaestra; some already existing palaestras were either built over by neighbouring structures or paved in stone changing them from exercise grounds to civic plazas.”174 Yet, sports and games were practiced even by Christians if Sidonius enthusiastically describes in considerable detail a ball game at the villa of a friend in which “players faced each other, caught and threw a ball, turned swiftly and ducked, dodged the ball, and used substitutes.”175 It is very likely that the porticoes and the central courtyard of the Villa at Palazzi di Casignana being near the baths were used for social gatherings in the manner of a ‘civic plaza’ and also for the practice of sport and games in the manner of a palaestra.

sequence of bathing. See for example Celsus, 2.17.1-10, 3.6.13-14, 3.12.3-4 and for Galen particularly De Sanitate Tuenda 6.1-452. Carcopino cites Martial as evidence for alternate bathing routes, see Mart., Epig. 5.42. CARCOPINO 1940, p. 261. 169 BALSDON, J.P.V.D. Life and Leisure in (New York 1969), p. 29; DUPONT, F., Daily Life in Ancient Rome, trans. C. Woodall (Oxford: Blackwell, 1992), p. 264. 170 Piazza Armerina: WILSON 1983; Patti:VOZA 1976-1977; Porto Saturo: LATTANZI 1973, p.45; Desenzano: SCAGLIARINI CORLAITA 1994, pp.52, 54-55. 171 WILSON 1983, fig.1. 172 BALMELLE 2001, passim. 173 For the condemnation of Greek methods of exercise under the Empire, see Pliny NH, 29, 26; 35, 168; Seneca, Ep. 88,18; Tacitus, Annales, 14, 20. CARCOPINO 1940, p. 257 with notes. 174 YEGÜL, F., “Bathing,” in G.W. BOWERSOCK, P.R.L. BROWN, O. GRABAR (ed.), Late Antiquity. A guide to the Postclassical World, 1999, p.338. 175 Quote from CROWTHER, N.B., Sport in Ancient times, 2007, p.159. 66

4.27. Building G (in green).

67

Roman Bathing in the fourth century AD

In the fourth century AD, private baths were a common feature of the elite residences all over the Empire. The massive structures, the scale of the rooms, and the opulence of the decoration of these facilities convey a symbolic message of wealth, luxury and aristocracy. Private baths likely emulated the grandiose public baths, such as the of Diocletian, those of Constantine (fig. 4.28),176 and baths built inside the imperial residences such as the Villa of Maxentius in Rome.177

4.28. Rome: The Thermae of Constantine (4th c. AD).178

In Spain, private baths are found in the villa at Torre de Cardeira, which include a series of polygonal rooms in linear arrangement.179 The villa at Milreu in Lusitania dated to fourth century AD also contains a huge thermal facility.180 In Aquitania there are several examples of private baths inside villas, usually located near the peristyle, and Sidonius Apollinaris describes a few in his works.181 In Bruttii, private baths have been uncovered inside several late imperial

176 YEGÜL, F., Bath and Bathing in , 1992; FAGAN, G.G., Bathing in Public in the Roman World, 1999; MERTEN, E. W. Bäder und Badegepflogenheiten in der Darstellung der Historia , 1983; WEBER, M. Antike Badekultur, 1996. 177 For the villa of Maxentius see PISANI-SARTORIO-CALZA 1976, p.128. 178After: http://images.google.com/imgres?imgurl=http://frysingerreunion.org/1/france/arles03.jpg&imgrefurl=http://www.ga lenfrysinger.com/roman_mosaics_arles.htm&usg=__sCWiOQ32Znga5fsv61ARzrYxTPE=&h=453&w=648&sz=12 0&hl=en&start=30&um=1&tbnid=- 0R38jggAEq13M:&tbnh=96&tbnw=137&prev=/images%3Fq%3DRoman%2Bmosaics%26ndsp%3D20%26hl%3D en%26rlz%3D1T4GGLJ_en-GBIT250IT250%26sa%3DN%26start%3D20%26um%3D1. 179 GORGES 1979, p. 475. 180 GORGES 1979, pp.132, 480-481. Idem 1990. 181 BALMELLE 2001, p.136, 187 ff. 68 villas some of which are treated in this study (Chapter 6). There is a remarkable bath at Bulla Regia in Tunisia, which features an octagonal frigidarium similar to Room 12 at Palazzi di Casignana. 182 Even the villa at Piazza Armerina (fig. 4.29) has an octagonal frigidarium (see room at the left lower corner of the plan).183 The baths of a fourth century Villa at Acconia di Curinga (fig.4.29), north of Vibo Valentia in Bruttii, truly stand out for monumentality and well built vaulted structures. Although they still stand in place (fig.4.30), they have never been fully excavated and published.184

4.29. Plan of 4th. c. private baths in Acconia di Curinga, Calabria (Italy).185

182 FAGAN 2005. 183 WILSON 1983, fig.1. 184 See ARSLAN 1966; ACCARDO 2000, pp. 167-173 for a very brief account. 185 Ibidem. 69

4.30. Extant structures of the private baths in Acconia di Curinga (4th. c. AD).186

186 After http://www.calabriaintour.it/images/termeacconia03.jpg. 70

4.31. Villa at Piazza Armerina (mid-4th c. AD).187

187 WILSON 1983, fig.1. 71

The diffusion of private baths during the late Roman period is linked to the decrease of financial availability of the impoverished municipia. Under Constantine and Costantius the Second, a general impoverishment of cities income occurred. Many services such as the baths were built by the cities relying mainly upon private financial contributions or through the intervention of decuriones and extraordinary forms of taxation.188 Thus, there was a drastic reduction in the number of public baths in comparison with the early imperial period. Noteworthy is that the doubling of baths within the same building, as we find at Palazzi di Casignana, is not attested in urban contexts.189 This can be explained with the lack of space inside towns, but also points to a likely decrease of municipal financial resources during the late imperial period. Private baths until the fourth century AD did not usually take up large space. Small baths were in fact less complicated to build with their heating system and doomed roofs. A small bath was also less difficult to keep hot. At Villa Palazzi di Casignana the combined size of the Western and Eastern Baths is approximately 1,100 square meteres, indicating that the owner could easily afford the expense for heating, water and maintenance.190 Because they needed a huge amount of water, fuel and work force in order to function properly, baths, especially those of grand scale, were a privilege of the very wealthy Romans. According to the cost of water was very high.191 When the aqueduct was not privately owned by the villa’s owner, water provision required a fee and therefore was available only to families that could afford the expense.192 A. Bouet mantains that the largest sector within a double private bath of a villa, which at Palazzi di Casignana corresponds to the Western Baths, was used by men, whereas the smaller by women.193 During the time of Martial and Juvenal, under Domitian, and still under Trajan, there was no formal prohibition of mixed bathing. Women who were against bathing with men could bathe in balnea provided for their exclusive use. But many women wanted to practice the sports that preceded the baths in the public thermae and chose to compromise their reputation and bathe at the same time as men.194 Between the years 117 and 138 Hadrian passed a decree mentioned in the Historia Augusta which separated the sexes in the public baths. But since their plan included only one frigidarium, one tepidarium and one calidarium this separation was achieved in time rather than in space by assigning different hours for the men’s and women’s baths.195 It is very possible that the doubling of the baths inside the Villa at Palazzi di Casignana was motivated by the same need of separating women from men when bathing.196 In one carmen Sidonius Apollinaris mentions the existence of ‘winter baths’ (thermis hiemalibus) inside the villa of his friend Pontius Leontius, which suggests that there were two baths, one used in the

188 JONES 1974, pp. 980-983. See also KELLY 2004. 189 As Baldini Lippolis points out, the House of Europe and that of Castorius at Djemila contain two baths but they were built in two different times. In Aquileia, private baths are not documented until the sixth century AD, perhaps because the public facilities were still functioning, BALDININI LIPPOLIS 2001, p.65. 190 See BALMELLE 2001, p.178 ff. for an analysis of the luxurious baths inside the villas of Aquitania to be compared with those at Palazzi di Casignana. 191 Frontinus, De Aqueductibus. HODGE, A.T. Roman Aqueducts & Water Supply, London: Duckworth (2001). 192 JONES 1974, p. 984. 193 BOUET 1997-1998, pp. 241-242. 194 CARCOPINO 1940, p. 258 refers to: Pliny, NH, 33, 153; Quintilian, 5, 9, 14; Martial, 3, 51 and 72; 7, 35; 11, 47; Juvenal, 6, 419. 195 CARCOPINO 1940, p.258 ff. with notes. 196 On mixed bathing see SCHÖLLGEN, G., “Balnea mixta: Entwicklung der Spätantiken Bademoral,” in Festschrift K., Thraede, 1995, pp.182-194. 72 winter and the other in the summer.197 Another explanation for the separation of two bathing suites and for the spread of private baths during the the late imperial period may lie in the new morality of the Christian faith, which encouraged people to preserve their modesty. The same Sidonius speaks of the custom of going with friends to private domestic baths in order to keep the pudor, or modesty, intact.198 A study by R. Ginouvès on the baths of fourth century Greece points to the increase in number of the tubs and the parallel reduction of their size, which indeed indicates a change of the bathing practice at this time.199 It is very likely that the construction of two baths at Palazzi di Casignana during the fourth century AD is linked to the distaste of the Christian faith for communal bathing, which is dramatically expressed by St. Jerome’s injunction: “He who has bathed in Christ, has no need of a second bath.”200

197 Sidonius, Carm. 22.180, I. 198 Sidonius, Carmen 23, 495-499; LIPPOLIS 2001, p.65. On bathing practices in public, see FAGAN 1992. 199 GINOUVÈS, R., “Sur un aspect de l'évolution des bains en Grèce vers le 4ème siècle de notre ère” in Bulletin de Correspondance Hellénique 79, 1955, p. 135-152; BALDINI LIPPOLIS 2001, p.65. 200 YEGÜL, F., “Bathing”, in G.W. BOWERSOCK, P.R.L. BROWN, O. GRABAR (ed.), Late Antiquity. A guide to the Postclassical World, p.338.

73

The architectonic changes in the Eastern Area of the Villa at Palazzi di Casignana between the third and fourth century AD.

4.32. Plan and ruins of the seaside, eastern area of the villa.201

201 After Guida, p.72, figg. 97-98. 74

The residential area of the villa lies at the eastern side of the complex at 300 meter distance from the sea. The extant ruins (fig.4.32) represent the result of a complex series of renovations and additions that occurred between the third and mid-fourth century AD. In antiquity one could enter this area from the eastern façade but also from Courtyard 55 (fig.4.12), which today is obliterated by the road SS 106. The structures visible today are situated between the train track and road SS 106 (see the road on fig.4.32). The latter was built on top of a large section of the villa’s seaside area. In this part of the complex, several rooms were decorated with floor mosaics.202

Corridor U (fig. 4.32) is a narrow corridor that runs east- west in direction of the baths at the south side of Area E. It was decorated with a monochrome mosaic of green tesserae.

4.33. Room R with emblema of the mosaic floor.203

Room R (fig.4.33). Near corridor U there is Room R which has a round plan, and had two columns not longer extant. Only one marble column base is preserved together with the lower parts of two pilasters that supported the roof (fig. 4.33). Circular rooms of small dimensions are often documented near large reception halls. They recur for instance in Rome at the House of the Seven Halls, in the Huilerie at Salamina of , and at the Villa Urbana in Sirmium.204

202 These rooms were built in slight different chronological phases and with different technique. See Guida, pp.82- 83. 203 After Guida, p.82, figg. 113, 114ab. 204 LIPPOLIS 2001, pp.63-64. 75

Room R at Palazzi di Casignana was a small lodge that was open to allow a view of the garden. The floor is decorated with a mosaic featuring a large band of vegetal spirals rendered in white and black tesserae on a green background. At the centre, the polychrome emblema in small tesserae represents an unidentified crowned female face that looks south (fig. 4.33).

4.34. Room O: mosaic floor with floral motif. 205

Room O (fig.4.34). This room is positioned at the western side of Room Q (fig. 4.28). It was severely damaged by the construction of the modern road SS 106. It contains portions of a floor mosaic framed by a large band with green tesserae. The motif is formed by large white flowers on a green background.

4.35. Room Q. Mosaic representing Dionysus and Satyr.206

Room Q (fig.4.35). It is adjacent to Room R but was built during a slightly earlier phase than Room R.207 Its floor mosaic represents Dionysus pouring wine into an amphora while he is assisted by a Satyr who stands behind the god. The Dionysian decorative repertory among the pagan motifs is frequently attested in late antiquity inside triclinia and reception rooms.208

205 After Guida, p. 82, fig.112. 206 After Guida, p.83, fig.116. 207 Guida, pp.75 ff. 208 Several examples are in BALDINI LIPPOLIS 2001, pp. 63-64. 76

The area between Room Q and Room R (fig.4.32) is decorated with a floor mosaic that features a motif with lotus flowers containing two spirals at the centre. This mosaic was intentionally placed to link Room Q to Room R, which were chronologically different. Room Q and R seem to create one space, a sort of elegant small belvedere on the south side of the villa. Room Q, at the east side, communicates with the apsidal Room T and Portico P; at the western side with Room O and northern side with Room M.

ROOM M (fig. 4.32). It is located at the northern side of Room Q. It features a mosaic with large white, black and green tesserae that create a motif of peltae. Room M communicates with Room N on its western side and Room Q on its southern side.

Rooms U-O-N (fig. 4.32) These rooms were damaged by the construction of the road SS 106. They were all paved with mosaics and decorated with painted frescos, fragments of which remain on the site. Each of these three rooms has a marble threshold. Although these rooms are not large, their sumptuousness suggests that they were used for reception or dining.

ROOM N (fig. 4.32). This room was partially damaged by the modern road. It features a floor mosaic with large white, black and green tesserae that create a motif with four petal flowers inscribed inside a losanghe. The mosaic is framed by a large band rendered with green tesserae. Room N is accessable through an entrance placed on the western side of Room M. On the northern side of Room N lies Room S that has been seriously damaged by road S.S. 106. There is no trace of the most recent phase of this room. The destruction caused by the road uncovered suspensurae under the pavement that belonged to an earlier phase of the room predating the fourth century renovation of the villa. Room S was likely heated in an earlier phase but the severe destruction of this space makes it nearly impossible to assess its role during the fourth century AD. Room I and its flanking squared Rooms H and L will be be described in Chapter 9. Room B (ca. 8.20 m. x 6.70 m.) also stands out for size and rich decoration (fig. 4.32). Its floor mosaic represents the Four Seasons, likely referring to a reception use of this room. The mosaic is described in Chapter 5. The following section will look at the seafront façade of the villa and discuss its reconstruction.

77

The Eastern Façade of the Villa at Palazzi di Casignana in the fourth century AD.

4.36. Reconstruction by V. de Nittis.209

The eastern façade of Villa at Palazzi di Casignana, which looks out towards the sea, has been reconstructed by V. de Nittis in two ways as represented respectively in fig. 4.36. 1) A long wall featuring a large entrance door at the centre, flanked by two towers and surmounted by a columned and arched walking gallery (top image), and 2) a portico with pilasters flanked by two towers and surmounted by a lodge with three columned arches (bottom image). In both restorations, de Nittis suggests that the eastern façade features a second story and two towers on the opposite ends. The upper level of the façade is reconstructed in two different ways: 1) with an arched gallery formed by 12 columns (bottom image) 2) an arched lodge with three columns (top immage). In both suggestions, the two towers placed at the end of the façade seem to be a distinctive and undeniable feature.210 Although these reconstructions are appealing, they are not convincing after taking a closer look at the surviving remains of the eastern façade. Only the ground level structures are preserved, including the lower portions of the portico’s walls and parts of its polychrome floor mosaic. The surviving structural elements are not sufficient enough and suitable to support de Nittis’ reconstruction of a second story in Area D of the villa. In the archaeological report by the villa’s excavators, it is specified that the foundations of the eastern façade were not deep.211 This element alone works against a reconstruction of towers and a second story. With low foundations, the wall of the portico could not have been sturdy enough to support the weight of an upper level. So far, only one column drum has been uncovered in the area of the eastern façade, which is too little evidence for reconstructing an upper floor that featured an arched

209 After Guida, p.112, fig.170-171. 210 V. DE NITTIS 2006; idem 2007, p.112-113. On towers represented on African mosaics see DUVAL 2002. 211 See Guida, p. 104. 78 gallery with 12 columns, or even an arched loge with three columns for that matter. It seems more likely that the eastern façade was designed with one story featuring a portico (Portico P) that measured 22.50 m. x 3.75 m. The portico was divided by seven pilasters as indicated by their bases, which likely were connected by arches at interval of 2.20 m. Frescoes of vivacious colours likely decorated the walls of the portico since many brilliant fragments of painted plaster were found in the fallen structures around the area. Even the reconstruction of the two towers at the opposite ends of the eastern façade, as attractive as it is to imagine, seems unrealistic. According to de Nittis, the apsidal shape of both Room A and Room T at the opposite ends of Portico P suggest that two towers were built on top of them. These would be similar to those of villas represented on the northern African mosaics dated to fourth century AD, such as those at Tabarka, Oudna, and Carthago.212

4.37. Buttresses of Room A.213

Besides the low foundations, towers seem unlikely for the following reasons:

1) Room A features very low reinforcing buttresses that lean against the exterior walls’ lower part (fig. 4.37). These walls are partly preserved and reach a maximum height of 3 m. The buttresses are not massive enough to support the weight of a taller structure like a tower. Even at the Villa of Piazza Armerina, similar buttresses were found, for instance, against the external

212 DE NITTIS 2006; for the African mosaics see DUNBABIN 1978, 121-123; SWARNOWSKI 1978, 61-64; DUVAL 1986. 213 After Guida, p.88, fig.126. 79 wall of the north apse of the Great Hunt Corridor.214 The buttresses likely served to contain the dirt external to the walls. 2) Room A featured three large windows that open at only 40 cm. from the ground level. Easily accessible, these windows do not seem compatible with the structure of a tower, even if one hypothesizes that the tower was built for an aesthetic or symbolic reason and not for a defensive purpose. More likely the windows were created to provide the viewer with a breathtaking panorama of the garden and the Ionian Sea in front of the villa. Room A was paved with a geometric and vegetal polychrome mosaic rendered in small tesserae. Only a portion of the mosaic still lies in place along the curved wall of the apse. The presence of an elaborate floor mosaic indicates that the room was likely roofed, possibly with a dome as it occurs in late Roman architecture at this time. Rather than a tower, it can be assumed that Rooms A and T were small dining rooms. There is evidence of small summer dining rooms featuring a rounded shape (cenatiuncula) in late Roman villas. J. Rossiter shows two examples. The first is represented by a small round room inside the villa at Almenara de Adaja in Spain that was roofed and used as a summer dining room. It was designed to contain a single stibadium from which the diners could enjoy a close view of the villa’s garden from a shaded space. The second example is provided by two pillared hemicycles that flank the inner peristyle of the villa at Montmaurin in Aquitania, belonging to a large summer dining facility.215 It is possible that Room A and Room T at Palazzi di Casignana, being both spaces open to the view of the garden and sea, were small but elegantly decorated lodges of Portico P where people could stop and enjoy the view of the garden and the sea. They might have been used as small summer dining rooms capable of receiving a stibadium for more intimate dinners. 3) Room A has a large entrance door paved with marble that opened at the centre of the apsidal eastern wall. Such an elegant entrance seems to be discordant with the architectural style of a tower. 4) Room T, which is on the north opposite side of the eastern façade, in symmetrical position in respect to Room A, likely featured the same characteristics as Room A, including three large windows. These openings, as it is the case for Room A, would call into question the reconstruction of a tower on the top. Very small portions of the walls in Room T still stand in place, and only a few, much damaged strips of the polychrome floor mosaic have been recovered. Also Room T was paved with geometric and vegetal mosaics rendered with small tesserae. Noteworthy is the presence of six steps outside Room T, placed between the buttresses of the apsidal wall.216 An exterior staircase would be another anomaly for a tower. Archaeologists have interpreted the steps as a way to even out a difference in ground level between the higher Room T and the garden outside. 217 This study suggests that the placement on a higher level of Room T and thus, of the entire sector wherein lies Room I (the largest reception room featuring an apse and a cruciform plan), may have been intentionally created to give prominence to the most prestigious area of the villa in which the dominus received his protected clients or prestigious guests and where ultimately he was glorified. A difference in

214 WILSON 1983, p.24. 215 ROSSITER J., “Convivium and Villa in Late Antiquity,” in Dining in Classical Context, ed. by Slater W.J., Ann Arbor 1991, p. 205 with notes. 216 Guida, p. 106,fig.161. 217 See Guida p.79. 80 ground level also occurs in the fourth century villa at Desenzano in northern Italy, right in the area where the grand apsidal reception room of the villa is located.218 In short, more architectonical elements from the ground are needed to provide an accurate reconstruction of the eastern façade of villa at Palazzi di Casignana. Yet, based on the archaeological remains, it is likely that the eastern façade of Villa at Palazzi di Casignana featured a walking belvedere flanked by two apsidal rooms. All three elements, Portico P and the two projections Room A and Room T, share a similar openness by means of windows and arches that aim at embracing an alluring view of the garden and the sea. It is inspiring to imagine the fourth century AD owners and guests of the villa walking along the portico paved with refined polychrome mosaic and decorated with brilliant coloured frescoes, while walking so near to the glimmering Ionian Sea. It is even appealing to envision how visitors approaching from the sea were awe struck by the grandiosity and magnificence of the Villa at Palazzi di Casignana. The monumental façade with a portico and symmetrical lodges by the sea must have been an incomparable sight to see. The search for striking views from the sea by Roman villa architects dates back to the Republican period, as witnessed by several Roman wall paintings that reproduced monumental facades of maritime villas along the Campania coast. Starting in the early Republican period maritime villas with long colonnaded porticus flourished along the Italian coasts, taking full advantage of sea views and rugged Italian shores.219 Villa owners tried to create places of beauty in which the natural topography of the coast and the architecture of the villa all blended together harmoniously. 220 The porticus, opening directly on to the shore, provided the owners and their visitors with a spectacular walking area (ambulatio). In the early imperial time, the maritime villa at Damecuta, perhaps built by the Emperor in , exemplifies this coming together of nature and artifice much wanted by Roman patrons and architects (fig.4.38). The villa lies on a plateau beneath Monte Solaro at the north-west corner of the island, and converges on the seaward through a splendid porticoed loggia-veranda featuring seating alcoves along the shore line, at approximately 492 feet above sea level. The west sea-facing side of the porticus features a central curved projection for ambulatio, while the eastern side features a veranda or belvedere.221 Recent excavations at Castellamare di Stabia (ancient Stabiae) have yielded extraordinary remains of villas dating between the Julian-Claudian and Flavian age with porticus –ambulatio, which are known also from Campanian wall-paintings (fig.4.39).222

218 SCAGLIARINI CORLAITA 1992, p. 28 fig.13, n.35. 219 For an account of the ancient sources that provide descriptions of porticoed villas on the Bay of Naples see D’ARMS 1970, p. 127 ff.; Statius (45-96 AD), Silvae, II.2 A Roman Seaside Villa. This is the first long poem devoted to the description of a villa. 220 The design of the colonnade porticus in Roman maritimae villae may have evolved from a long narrow row of rooms opening on to a road or a court, as it is still visible in the Near East, or could have originated from the Greek stoa, the Hellenistic shopping center. See McKAY 1975, p.117 and bibliography of note 192; SWODOBA, K.M., Römische und romanische Paläste, 1919. 221The veranda-belvedere is surmounted today by a medieval tower and by an isolated suite. This contains small but elegantly furnished rooms built into the Cliffside which are accessible only by a narrow ramp and steps. The difficulty of access and the similarity of the belvedere to that of the Villa Jovis suggest that the residence may have been designed for Tiberius. See McKAY 1975, pp.116-118. A recent account of the site is in Krause C., Villa Jovis : die Residenz des Tiberius auf Capri, : Von Zabern, 2003. 222 For the villas at San Marco di Stabia see LONGOBARDI 2006. A rich bibliography is in McKAY 1975, p.248, note 196. 81

4.38. Villa at Damecuta, Capri (1st c. AD).223

223 After McKay 1975, p.116, fig.44. 82

4.39. Villa S. Marco at Stabiae (1st c. AD).224

During the second century AD, an external portico on the main façade of the villa, like the one in Palazzi di Casignana, is found in Aquitania at the Villa of Plassac.225 Even in the fourth century AD, many peristyle houses are elongated only on one side by a belvedere gallery. For instance, the oriental façade of the villa at Lalonquette in France dated to the fourth century features a gallery running parallel to the seashore with two opposed apses (fig. 4.40). There, the portico seems not to have had pilasters or columns like those found at Palazzi di Casignana. Other parallels may be established with the house at Hippone in Algeria226 and with the villa at La Olmeda in Spain (fig.4.41).227

224 After MCKAY 1975, p.118, fig.45. 225 BALMELLE 2001, p. 130 note 189. 226 See, for instance, the house in front of the sea in LAVERGNE, D., Le quartier des ‘villae du front de mer,’ pp. 127-138. 227 CHAVARRIA-ARCE-BROGIOLO 2006. 83

4.40. The fourth century Villa at Lalonquette in Aquitania (top); and the fourth century Villa at Carranque in Spain (bottom).228

228 After BALMELLE 2001, p.129. 84

4.41. Villa at La Olmeda, Spain (4th. c. AD).229

In other instances, the portico is inserted inside the villa’s plan as it occurs at the villa of Piazza Armerina230 and at the villa of Carranque.231 In the last two cases the porticos represent the most prestigious areas of the villa. A portico, whether just one or two, typically highlights the main entrance and provides a panoramic belvedere. In the villa of Valentine in Aquitania and in the late Roman villa of La Sevillana, the portico functions as a monumental entrance to the residence.232 At the fourth century villas at Seviac, Mocrabeau-Baptest, Saint-Cricq-Villeneuve,

229 After BROGIOLO-CHAVARRIA 2005, p.42, fig.18. 230 The portico there is 68 m. long; see CARANDINI et al. 1982, pl. III, no. 36, p.194 ff. 231 The portico is 38 meter long, see FERNANDEZ-GALIANO 1994, p. 266, fig.1. 232 For the Villa of Valentine see BALMELLE 2001, p.148-149; for the villa in La Sevillana, see AGUILAR SÁENZ & GUICHARD 1993, pp. 116-123, fig. 38. 85 and Lalonquette in Aquitania, the portico looked out towards a remarkable landscape of water that embellished the valley offering a pleasant and relaxing vista.233 The comparison with other fourth century villas discovered in other parts of the Empire shows that even at Palazzi di Casignana there was a taste for impressive views, and this motivated its owners and architects to create a façade offering a spectacular panorama. The combination of peristyle plan with a portico façade is a characteristic shared with several late imperial houses. Particularly, the portico with symmetrical apsidal rooms on the short sides is a distinctive feature of the wealthiest and most luxurious villas across the Empire.234

233 BALMELLE 2001. 234 Ibidem, p.130 ff. For examples of villas featuring a gallery/portico with projecting ends on both sides see Swodoba 1969, 77-132; At Moncrabeau-Bapteste, the north façade with a portico featuring two angled spaces recalls the plan of portico villas “mit Eckrisaliten” even if in this case there are no projections. 86

CHAPTER 5: The Mosaics of the Villa at Palazzi di Casignana.

Introduction

In Roman houses the decorative apparatus formed an ensamble with the domestic architecture. Mosaic floors, paintings, architectural ornamentations, and statuary reveal much about the wealth and social aspirations of the house owner. Res publica and res privata were intertwined inside the Roman household and their boundaries were blurred in Roman life.235 The dominus, the head of the household, performed a large part of his social and public life within the private sphere of his residence. The house was treated as a public stage on which one’s social, political and familial duties were performed accordingly to one’s background. The copious display of imagery, rich materials, and luxurious furnishings was intented to celebrate the owner’s culture, personal history and public life.236 Therefore, the Roman house becomes an important document for social historians as it speaks about the rank, individual achievements and class ambitions of its patron-owner. Most of our evidence of Roman domestic life is drawn from the houses of the middle and upper classes and little is documented about the lower strata of society. The study of Roman domestic architecture, especially that of lavish villas, largely focuses on the most attractive remains leaving often the utilitarian quarters of servants and slaves unexplored or little examined. Among the various forms of house decoration, mosaics represent the richest evidence for social studies. They provide a kind of guide for the modern visitor to the different parts of the house. By means of specific indicators, such as the subjects of figural compositions, the complexity or simplicity of geometric patterns, the diverse use of colors, and display of technical ability, mosaics invite the modern viewer to share some of the experience contrived for guests, peers and clients who regularly visited the house. Mosaics carry information about the house patron in a language that is often direct and not difficult to interpret, especially for those familiar with ancient literary texts. Some figural representations, such as the ones that relate to the Greek theater or mythological stories, for instance, remark the classical cultural background of the house owner. Mythological scenes or characters were deliberately inserted more than once on the same floor or within the same house to reinforce a message of economic strength and social power. Often too mosaics help to distinguish the function of spaces within the house. The mosaics at Palazzi di Casignana offer an interesting case for the study of the Roman villa as a favored arena for the social and cultural life led by the Roman elite of Bruttii during the late imperial period. They represent the most attractive surviving feature of the villa today, producing a striking impact on the visitors for their sheer quantity more than for their artistic quality, which at times is badly low. The main task of this chapter is to present a catalogue of the villa’s mosaics addressing their stylistic, iconographical, technical, and chronological characteristics. This has been possible because the pavements are preserved in their original setting and in an excavated site that allows us a reconstruction of a large part of the villa’s floors. For each mosaic, this catalogue provides a brief description of its location, dimensions, material, a probable date, the geometric and figural motifs, and a few parallels with other floors found all over the Empire. I have drawn this

235 See WALLACE-HADRILL 1988 for a thoroughly treatment on the subject. Idem, 1998. See also CLARKE 1991. On Roman villas and their social implicatios see SMITH 1997; HALES 2004. 236 On the opulent display of decorations inside a fourth century villa such as the villa at Piazza Armerina, see WILSON 1983b. 87 information manly from personal notes and photographs taken during several visits at the villa, which the Archaeological Superintendence of Calabria permitted me to undertake from 2007 onwards. This inventory has to be taken as a starting point in the study of the floors inside the Villa at Palazzi di Casignana. Some details such as, for instance, the density and exact size of the tesserae, the dimensions of extant floors have not been reported with accuracy on account of the impossibility of accessing several rooms of the villa, which are currently under restoration (the whole site is still closed to the public). I am indebted to the work of the archaeologists who excavated the villa and provided a preliminary description of its floors in the report that they published in 2007.237 There is still much to enquire about the mosaics of the Villa at Palazzi di Casignana. In particular we must ask whence the inspiration for the motifs adopted came; how many were the craftsmen responsible for the whole flooring; were all the mosaics the product of a single workshop? or –as it more probable, the work of several mosaicists active at the site during the same or different periods?. The state of our knowledge about the organization of mosaic craftsmanship in antiquity is still inadequate for us to provide definitive answers to these and similar questions. Moreover, a comprehensive study of the Roman mosaics in Bruttii has not been yet undertaken, making it really difficult to attempt final conclusions. There is not sufficient evidence of Roman mosaics from Bruttii to allow us to identify local workshops that were capable of satisfying the extraordinary commissions made for the Villa at Palazzi di Casignana. The existence of a few exceptional artworks in Bruttii during the late imperial period, such as the third century sarcophagus of , speak of very wealthy patrons who were eager to furnish their homes with expensive and elaborate works of art. But there is little to suggest that these works were locally made. Regardless of the blurry picture on the Roman mosaic art in Calabria, it can be said with some confidence that all the floors that decorate the Villa at Palazzi di Casignana find close parallels more or less in some part of the Empire or other. This, in turn, would confirm that the mosaic workshops (officinae) all over the Roman world used copy-books (whatever form they took), which contained a collection of motifs and patterns that circulated widely among mosaicists. These craftsmen did not create a wholly original floor but most likely drew their designs from a shared repertoire while creating personal variations of popular patterns.238 At Palazzi di Casignana, it seems likely that several different designers drew the decoration of each floor in close consultation with the villa’s patrons. This expensive collection of choice mosaics inside one building stands as remarkable evidence that during the late imperial period cultured and refined elite lived in the Locri’s area. The dominus of the Villa at Palazzi di Casignana expressed his wealth and taste by means of a medium that in the Roman world had been for centuries associated with luxury, opulence and high social status.

237 C. SABBIONE (ed.), La Villa Romana di Palazzi di Casignana. Guida Archeologica, 2007. 238 On this subject see DUNBABIN 1999, pp.269-278. 88

CORRIDOR 19

Cat. No.1 (a). Corridor 19. Fragment of mosaic with a shield motif forming a cross.239

Site Reference:240 Corridor 19, threshold that leads to Room 12.

Dimensions: ca. 90 cm.

Material: small tesserae of black and white limestone.

Date: fourth century AD.241

Description: motif of crosses formed by white outlined scuta (shields) placed around a square that contains a floret. Each scutum (shield) is adorned with a short-lobed ivy leaf with volutes (Cat. No. 1 a)

Observations: no comparative examples have been found, and it is probable that this pattern was an independent creation of the mosaic artist. The date of fourth century AD is based on the stratigraphic data of Corridor 19. 242

239 After Guida, p.50, fig.53. 240 For the location of each mosaic in this catalogue see fig.1.5 (the baths) and fig..4.32 (the seafront area of the villa). 241 This corridor was built during the T8 phase of the villa: Guida, p. 105. 89

Cat. No.1 (b). Corridor 19. Mosaic with motif of peltae.243

Site Reference: Corridor 19.

Dimensions: not determined.

Material: small tesserae of black and white limestone.

Date: fourth century AD.244

Description: motif of tangent pairs of backed, black peltae, alternately upright and recumbent on a white background (Cat. No. 1 b).

Observations: The motif of peltae is quite diffused on Roman mosaics. It is found from the first century BC onward all over the Empire in different variations. See, for instance, inside the Villa at Piazza Armerina,245at Carthage (Tunisia)246, Djebel Oust (Tunisia),247 and Antioch ().248A close parallel to the motif of Corridor 19 (Cat. No. 1 b) is found in Koln, (), 249and at Italica in Spain.250

242 For the scuta (shield) design see Décor, I, p.237 e. 243 After the brochure: La Villa di Palazzi di Casignana. Progetto scavo aperto. Visita al Cantiere di Scavo Archeologico. Mostra Documentaria e Multimediale 1-31 Agosto 2005, p.8. 244 Guida, p. 105. 245 BAUM 2003, p.196, pl.46.1. 246 GAUCKLER 1913, Basiliques, pl.1. 247 CMGR I, p.173, fig.16.3. 248 LEVI 1947, pl.37, a. 249 PARLASCA 1959, pl.62, 2. 250 DUNBABIN 1999, p.170, fig.24. 90

ROOM 12

Cat. No. 2 (a). Mosaic of Room 12 with a motif of plastic cubes.251

251 After Guida, p.51, fig.54. 91

Cat. No. 2 (b). Detail of the prospective cube motif.252

Site Reference: Room 12, frigidarium of the Western Baths.

Dimensions: ca. 6 m. x 6 m.

Material: limestone small tesserae (1 cm) in black, white and green color.

Date: end of the third-early fourth century AD.253

Description: The central field of the floor features green and white diamonds that form three- dimensional cubes (Cat. No. 2 b). This motif is widespread from the second century BC onwards. It appears on an opus sectile floor in Pompei inside the tablinum at the House of the Faun (ca. 100 BC)254 and in Sicily on a mosaic inside the House in Piazza Vittoria in Palermo (fig.5.1.), and at the House of Leda in Solunto.255 The illusionistic effect of perspective cubes does not impose a specific orientation to the viewer. Wherever the viewer stands inside Room 12 he or she can fully appreciate the simple yet striking effect of this floor’s decoration.

252 Photo by author. 253 Guida, p.47. 254 LING, R., Ancient Mosaics, 1998, p.33, fig. 19. WESTGATE, R., “Pavimenta Atque Emblemata Vermiculata: Regional Styles in Hellenistic Mosaic and the First Mosaics At Pompeii,” American Journal of Archaeology 104 (2), 2000, pp. 255–275. 255 Von BOESELAGER, D.V., Antike Mosaiken in Sizilien, 1983, pl. 10, no. 20 (House in Piazza Vittoria, Palermo); pl. 31 (House of Leda in Solunto). 92

Fig. 5.1. Mosaic at Piazza della Vittoria in Palermo, Sicily (2nd c. AD).256

The framing band of Room 12 (Cat. No.2 c) features a white background with black squares of swastika-meander alternated with black squares, which are inscribed with black circles. The meander pattern appears during the Archaic period and continues to be used until the late imperial times with different variations. At Rome it is found, for instance, inside the “House of Catilina,” at the House below the Lararium, and in Horace’s Sabine Farm.257 It also recurs in Balazote, Spain.258 In Room 12, the swastika-meander motif separates the four apses of the room from the central field of its floor. This may indicate a separate function of the apses within Room 12.

The apses (Cat. No. 2 d). The apses are decorated with a pattern of a half-shield of bipartite scales in contrasting colors (here black and white), which form 17 rows.259 This ‘peacock motif’ was used to decorate mainly thresholds and small areas during the second century AD but in a later period it was adopted to decorate entire floors.260 According to M. Blake, the scale pattern derived from metal-work.261 It is found in different contexts and in different periods in Ostia,262inside the villa at Piazza Armerina,263 in (Sicily),264 and in Timgad, Algeria.265

256 http://www.thejoyofshards.co.uk/visits/sicily/vittoria/ 257 For these three parallels see OVADIAH, A., Geometric and Floral Patterns in Ancient Mosaics, 1980, pp.100- 105. 258 Décor, p. 228; Corpus España VIII, n.33, pl. 29. 259 Décor, II, p. 140. 260 BECATTI 1961, pp.293-284. 261 BLAKE, MAAR, 8, 1930, p.82. 262 BECATTI 1961: pp.21-22, n.30; p.12, n.2; p.176 n.324. 263 BAUM 2003, II, pl. 46 nn.1-2 (Room 11 c). 264 Von BOESELAGER 1983, pl.67. 265 Décor, II, p. 140; GERMAIN 1969, Timgad, n.98, pl. 35; n. 119, pl. 39. 93

Cat. No. 2. (c). The framing band of Room 12 near tub 21.266

Cat. No. 2. (d). Mosaic of the apses with a peacock motif.267

266 Photo by author. 94

ROOM 9

Cat. No. 3 (a). Room 9 (frigidarium of the Eastern Baths). Mosaic with Nereid riding a lion.268

267 After Guida, p.52, fig.56. 268 Photo by author. 95

Cat. No. 3 (b). Nereid riding a sea-monster.269

Site Reference: Room 9, frigidarium of the Eastern Baths.

Dimensions: 7.20 m. x 6.75 m.

Material: green limestone of various shades and white marble. The size of the tesserae is 1.5 cm. The density of the tesserae per each area of 10 square centimetre is 28 on the border band; 45-47 on the sea background, Nereids and marine monsters; 60 on the face of one Nereid. 270

Date: The mosaic is dated to the first half of the third century AD.271

269 After Guida, p.56, fig.65. 270SABBIONE et Alii 1997;BARELLO-BRIZZI-CARDOSA, in Atti dell’Associazione per lo Studio e Conservazione del Mosaico Antico, (Palermo 9-13 dicembre 1996), Ravenna, pp. 388-390; Guida, p.45. 96

Description: This mosaic was severely damaged when changes were made inside the room during the later phases of the villa’s inhabitation. The south-eastern corner was occupied by a small space (2.70 m. x 2.15 m.) of unknown function, which obliterated part of the figural scene. Also the insertion of a small oval tub (21) that was placed between Room 9 and Room 12 damaged part of the tiger. In addition, the construction of two drainage conduits ruined the images of the lion and the bull. The mosaic’s band on the southern side of the room was also damaged. The need of providing Room 9 with a second tub, conduits and the small space of unknown function, had the priority over the preservation of the Nereid mosaic, which did not show its finest form during the latest phases of the villa’s occupation. The mosaic covers Room 9 entirely and is framed by a double band of white tesserae. The are rendered in white marble tesserae with green internal details. They are drawn clockwise along one side of the room and exhibit the ‘silhouette style’ that was introduced during the Severan age and widely represented on black and white figural mosaics of Ostia and Rome.272 The sea is rendered by large dark green tesserae of limestone (2 cm.) of different shades that create a strong contrast with the white marble tesserae of the Nereids and marine beasts. The waves are drawn with short white parallel lines placed underneath and at the side of each figure. This is a characteristic that is found frequently on African mosaics and that was taken up on Italian floors.273 The placing of each Nereid at a different corner of the room takes into account the movement of the observer inside the room and his or hers different point of observation.

The sea monsters: On the northern side of the room a Nereid rides a lion (Cat. No. 3 a), on the southern, side a second Nereid rides a horse. Both animals appear in profile with the head turned towards right. On the eastern side a Nereid rides a bull, on the western a tiger. The bull and the tiger are depicted in profile while turning their heads to their left. All four animals have the head, body and anterior legs typical of their species. On the contrary, their back features a snaky tail ending with three fins in a typical manner of sea-monsters. The image is flattened by schematization although the interior anatomical details are rendered with vigour by dark green lines characteristic of the silhouette style. These lines produce a movement of the figures even even if they are sketchy and unrealistic.

The Nereids:274 The nymphs are naked above the waist and show no ornaments. The bust and face are drawn frontally while the legs wrap around the animal that they are riding. One hand leans on the animal and the other is lifted to hold a ivy leaf or the hanging of the drape. The gestures are little theatrical, yet inert. All four Nereids show the same hairstyle with a split on the forehead. The copybook used by the mosaicists was likely one since the same scheme is repeated four times with a few variations. The head, facial features and limbs of each Nereid are outlined in green tesserae in the silhouette style (Cat. No. 3 b). Green internal lines mark the body divisions and the muscular volume in a sketchy way especially on the lower abdomen, upper chest and neck. The eyes are depicted with heavy lids and brows, which produce a fixed

271 Ibidem. 272 CLARKE 1979, pp. 87 ff. 273 DUNBABIN 1999, p.140. 274 The Nereids are sea nymphs, the fifty daughters of Nereus and . They often accompany Poseidon and are always friendly and helpful towards sailors fighting perilous storms. The most notable of them are Thetis, wife of Peleus and mother of Achilles; Amphitrite, wife of Poseidon; and Galatea, love of the Cyclops Polyphemus. See MORFORD, M.P.P –LENARDON, R.J., Classical Mythology, 1999, pp. 98-99. 97 glance toward the observer. The most continuous line is the one that defines the hip and the pectoral muscles. The width of the internal lines varies from the very thin details of the face to the heavy delineation of the abdomen. The drapery of the nymphs is rendered by a series of linear folds that cover the lower part of their body. The tesserae that make up these lines and accents are bisected, quartered and often shaped into tiny triangles to produce a volumetric effect. Elsewhere on the main field of the floor, tesserae are usually large but coarsely cut. The overall impression is that of a pictorial, vivid and engaging composition.

The arrangement of the figures: In Room 9, the composition of the Nereids placed at different angles of the floor and depicted freely on a dark green surface encourages the spectator to move around the room. The floor surface is fragmented into isolated images and provides several viewing points. Wherever one stands it is possible to see one or more of the figures to good effect, knowing that the other figures are better appreciated from another position. This kind of setting produces a better effect than the Hellenistic central emblema, which imposes a single optimal viewpoint. In spaces of circulation as baths, subjects such as sea-monsters ridden by nymphs, dolphins ridden by Cupids, and Neptune driving a group of sea-horses, typically privilege the multi-view scheme as these figures can be seen floating and moving on the mosaic’s continuous surface that resembles the sea. If Room 9 was vaulted, as it has been suggested, the Nereids’ mosaic with its large scale composition and sense of movement well matched the imposing ceilings and its large space.

Observations: This arrangement and silhouette style of this mosaic resemble that of the mosaic found inside the Terme Marittime at Ostia, which is dated to the beginning of third century AD.275 Based on the stratigraphy, this mosaic can be dated to the first half of the fourth century AD but stylistically it could belong to an earlier period. The marine thiasos of the Nereids is appropriate for a thermal facility and recurs very often all over the Empire. The nude or half- draped Nereid riding on the back of a sea-monster with a drapery over her head was frequently represented in Greek and Roman art. Its imagery was adopted throughout the history of mosaic art.

275 BECATTI 1961, Ostia, III, 2, Terme Marittime. On the iconography of sea gods, see BECATTI, G., Ninfe e divinità marine: ricerche mitologiche, iconografiche e stilistiche, 1971. 98

5.2. Tunisi, Bardo Museum. The “Nereid Garden Room” in Carthage (3rd c. AD).

A Nereid riding side-saddle on the back of a Ketos or sea-monster is found, for instance, in Carthage on a polychrome and well executed mosaic dated to the third century AD (fig.5.2). Several mosaics with Nereids riding a sea-monster are diffused especially during the fourth century AD in the form of the triumph of the marine Venus.276

276 For examples see DUNBABIN 1978, pp.154 ff. 99

ROOM 24

Cat. No. 4 (a). Mosaic of Room 24.277

277 Photo by author. 100

Cat. No. 4 (b). Mosaic of Room 24 with detail. Notice the marble veneers on the walls.278

Site Reference: Room 24, calidarium of the Western Baths.

Dimensions: ca.3 m. x 5 m.

Material: limestone black and white large tesserae.

Date: end of the third-early fourth century AD.279

Description: on a white background are inserted squares outlined with two black frames (Cat. No. 4 a and b). The squares contain a ziz-zag, black diagonal square inserted with a white cross

278 Photo by author. 279 Guida, p.47. 101

at the centre. Each side of the squares is adjacent to an elongated hexagon outlined in black and filled with a black lotus motif. The space in between is filled with lozenges and crosses both outlined in black. Each arm of the cross contains two superimposed black triangles and a black diagonal cross at the centre. The hexagons are inserted with black flared lotus with central lobe, which form two white brackets.280 The ‘pavonazzetto’ grey marble veneers on the walls perfectly match the black and white geometric motif of the floor.

Observations: A variant of this motif is found in Djebel, Tunisia,281and a more complex one recurs in Tunisia at Thuburbo Maius.282 The black and white geometric motifs such as this of Room 24 developed in Italy during the late Republic and early Empire but lasted for centuries. They were cheap and easy to create and for this reason they were widely adopted to decorate small and large floors.

280 Décor, II, p.48. 281 Décor, I, pl. 179 e; CMGR I, p. 173, fig.7. 282 Décor, I, pl. 252 e; LÉZINE 1969, pl. VII. 102

ROOM 36

Cat. No. 5 (a-b) Polychrome mosaic of Room 36 (calidarium).283

283 Cat. No. 4(a): Guida, p. 47. Cat. No 5 (b): Photo by author. 103

Site Reference: Room 36, calidarium of the Western Baths.

Dimensions: ca.4.5 m x 4.5 m.

Material: limestone; polychrome small tesserae.

Date: end of the third-beginning of the fourth century AD.284

Description: The frame of the floor features a pattern with eight pointed white stars outlined in black tesserae containing two concentric black circles. Between the stars there are lozenges. The floor’s central field is decorated with a centralized, wreath-like pattern. In a square and around an octagon flanked by 8 rectangles perpendicular to the diagonals and to the medians, of 8 half-stars of eight lozenges, contiguous by the points to each other and to the central octagon. The lozenges are set around the rectangles. They are filled with four-petal and form squares that contain a Solomon knot, rectangles filled with a guilloche, and triangles at each corner of the carpet. The octagon is framed by a dentil motif and is emphasized by a circle of containing a circle with a polychrome rosette in green white, and red tesserae.

Observations: parallels of this pattern are found in Sfax, Tunisie,285 on a mosaic in Vienne (France), 286 and in Piazza Armerina.287 The border-patterns, the geometric schemes and the central rosette are executed with a few colors that are blended to create a refined polychrome effect. The central rosette is emphasized by a vegetal motif. Similarly, several floors of the late antique period all over the Roman world feature vegetal or floral designs combined with geometric patterns.288

284 Guida, p.47. 285 FENDRI 1963, Sfax, pl. IV; Décor, p.122e; p. 123e. 286 Recueil , III, 2, n.265, pl. XXI; Décor, p. 394. 287 BAUM 2003, plates 56/4 and 78. 288 DUNBABIN 1999, pp.296-297 for a discussion on the floral and vegetal designs. See also KHADER 2001. 104

ROOM 29

Cat. No. 6. Room 29. Mosaic with floral motif.289

Site Reference: Room 29, calidarium of the Western Baths. The room is small and has an octagonal plan. Its western side is adjacent to Room 36. The mosaic is severely damaged and was covered by an opus signinum during the latest phase of the villa. Numerous glass tesserae found in the fallen structures of the room indicate that the ceiling was decorated with a mosaic.

Dimensions: ca. 40 cm x 50 cm.

Material: Limestone.

Date: beginning of the fourth century AD.290 The mosaic predates the opus signinum floor.

289 Guida, p.55, fig.62. 290 Guida, p.55. The T8 phase was dated to the fourth century AD: Guida, p.105. 105

Description: Polychrome mosaic made of small tesserae (Cat. No. 6). It is an orthogonal composition of intersecting circles, forming saltires of hollow spindles and concave squares srepresenting a motif of white four-petal flowers interlaced with white circles.

Observations: this pattern is found across the Empire in several variations. It recurs at Reims (France), El Jem (Tunisia) and Timgad (Algeria).291

ROOM 20

Cat. No. 7 (a). Mosaic of Room 20, tepidarium of the Western Baths.292

291 All three examples are cited in Décor, I, p.370 b. For the mosaic in Reims see RecGaule I, I, n.3; for the motif in El Jem see FOUCHER 1961, Thysdrus, pl.H.b; for the one in Timgad: GERMAIN 1969, n.195, pl.66.

292 After Guida, p.53, fig.58. 106

Cat. No. 7 (b). Detail of the frame.293

Site Reference: Room 20, tepidarium.294

Dimensions: ca. 5 m. x 4 m.

Material: black, white and green limestone.

Date: second half of third century AD.295

Description: The central field (Cat. No. 7 a) features a three-colored orthogonal pattern of spaced swastika-meander with single return in double filets, the spaces staggered and containing a square.296 The frame of the floor (Cat. No. 7 b) is decorated with a bichrome (black and green) ashlar pattern, or isodomic motif.

293 Ibidem. 294 The presence of tubuli only at the corners of the room suggests a use as a tepidarium. Guida, p.47. 295 Ibidem. 296 Décor, II, pl. 190, a; e. 107

Observations: The central field motif of this floor is found in Sarande, Albania297 and in Sousse, Tunisia.298 According to Ovadiah, all the meander patterns evolved from ths swastika motif and has its origin in architectural decoration.299

ROOM 25

Cat. No. 8 (a). Room 25 and Tub 26 (frigidarium of the Eastern Baths).300

297 ANAMALI 1974, Mos. Albanie, p.44. 298 FOUCHER 1963, Maison des Masques, fig.6, p. 9. 299 OVADIAH 1980, p.100. 300 After Guida, p.57, fig.68. 108

Cat. No. 8 (b). Room 25: opus sectile floor.301

Site Reference: Room 25, frigidarium of the Eastern Baths.

Dimensions: ca. 6.50 m. x 3.50 m.

Material: marble (white, light pink, gray and purple).

Date: first half of the third century AD.302

Description: This opus sectile floor (Cat. No. 8 a-b) belongs to the type that Guidobaldi defines the ‘modulo medio quadrato’ (medium squared module) of opus sectile pavements. Guidobaldi classifies a large, medium and small group based on the size of the marble tiles. The medium module is composed of squared tiles measuring between 30 and 90 cm. and never exceeding four Roman feet (equivalent to 120 cm. circa). 303 In Room 25, the first squared tile contains a smaller diagonal square, which is inserted with a third orthogonal square.304 The repetition of the same scheme for the entire pavement by means of squared tiles made it easier for the craftsmen to execute a precise work, in the manner of modern tiles. The frame of the entire floor is composed of rectangular tiles of purple marble.

301 Photo by author. 302 This pavement belongs to the same phase of construction as Room 7 and Room 20. See, Guida, p.47. 303 GUIDOBALDI 1985. For an overview of opus sectile see DUNBABIN 1999, pp. 254-268. 304 Ibidem, pp.61-62. figg 11a, 12 b. 109

Observations: The squared module is the most diffused on opus sectile floors in Rome during the third and fourth century AD for its simplicity but it also appears in many variations.305 The squared tile may contain disks, triangles, stars or points in various combination, repeated in contrasting colors.306 The majority of opus sectile floors across the Empire feature a repetition of the same scheme for the entire surface. The basic shapes of the marble tiles include squares (the most common), rhombi, triangles and hexagons. These may be combined to form a more complex design. Reuse of marble pieces in order to create opus sectile floors is also frequently attested during the Late Empire. It is recognized by the different thickness of the tiles and occasionally by marks including inscription on the reverse.307 Room 25 features round tiles that were reused and several modules present a difference in thickness. The marble used for opus sectile floors varies in type and colors. The most attractive designs are those created with contrasting colored marbles as, for instance, the one found in Ostia at the House of the Nymphaeum in which the orange giallo antico edged with green porphyry stands out against lighter colored marbles.308 The development of opus sectile floors is not clear since many designs lasted for centuries and the tiles often came from reused marble. In late Republican and Early Augustan period a large number of opus sectile floors feature simple designs created with basic geometric tiles of medium size that were repeated for the entire field or used in different combinations.309 Often geometric compositions in opus sectile technique are used in emblemata inserted in opus signinum or tesselatum pavements, a tradition that lasted in the Campanian cities until the mid-first century AD.310 From the time of Augustus onward, the richest houses feature opus sectile covering entire floors while the technique established itself in standardized types.311The use of large marble pieces was introduced during the first century AD in the richest and most prestigious residences. The aula porticata of Nero’s Domus Aurea (64- 68 AD) contains a very elaborate opus sectile floor of large module with a base of circa 1.5 m., rendered with squares that contain disks and framed by undulating bands. The space between the squares is filled with foliate forms finely executed.312 The marbles used include the purple and green porphyry, giallo antico and pavonazzetto. Another imperial residence, Hadrian’s Villa at Tivoli, features an extensive use of opus sectile floors in several rooms underlining their representative and public role within the complex. 313 The development of opus sectile floors during the second and third century is little known. 314 Generally, the standard geometric and medium modules remain in use. In the fourth century a revival of elaborate designs occurred by means of polychrome marble tiles, while the execution of the technique became less accurate compared to the earlier floors.315 Elaborate

305 Ibidem, pp.62-63, figg. 12-13. 306 DUNBABIN 1999, p. 260, fig.274. 307 Ibidem, p.256. 308 Ibidem, pp. 260-261; BECATTI G., Scavi di Ostia 4, Mosaici e Pavimenti marmorei (Rome 1961), 103-4 no. 189, fourth century AD. 309 For Pompei, see GUIDOBALDI F., OLEVANO F., TRUCCHI D., ‘Classificazione preliminare dei sectilia pavimenta di Pompei,’CMGR VI, pp.49-61; for an overview of opus sectile floors DUNBABIN 1999, p.254 ff. 310 DUNBABIN 1999, p.259, note 24. 311 OLEVANO F., GUIDOBALDI F., ‘Affermazione dei pavimenti in opus sectile in redazione marmorea,’ CCARB 40, 1993, pp.163-174. DUNBABIN 1999, p.259 note 23. 312DUNBABIN 1999, p.260, fig. 276. 313 GUIDOBALDI F., Sectilia Pavimenta di Villa Adriana, (MosAntIt, Studi Monografici, Rome 1994); DE FRANCESCHINI, M.,Villa Adriana. Mosaici-pavimenti-edifici, Rome 1991, pp.610-613. 314 For a treatment on the late antique opus sectile floors of Southern Italy see BALDINI LIPPOLIS 1996. 315 DUNBABIN 1999, pp. 260-261. 110

designs are found, for instance, in the wealthiest houses at Ostia, inside the House of the Nymphaeum316 and the House of Amore and Psyche both dated to the fourth century AD.317 There was a great variety of marbles especially in Rome because of the exotic marble trade that developed in this city to meet the large demands of the emperor and the aristocratic and wealthy patrons. The most predominant marbles were orange giallo antico, green porphyry among a variety of lighter coloured marbles.318 In Late Antiquity the decoration in opus sectile became very popular. The technique was often careless in comparison to earlier times. For insance, the slabs were not cut but worked with a pointed hammer and often the edges of each tile were left unpolished. According to L. Hansen, this sloppiness may be a trend of the time. The house owners were apparently more interested in having a huge variety of colored marbles, often very rare, on their floors rather than an accurate execution of the opus sectile floor.319

316 BECATTI, G., Scavi di Ostia 4, Mosaici e Pavimenti marmorei, Rome 1961, pp. 103-104, no. 189, pl. CCVVII, fourth century AD. 317 BECATTI 1961, pp.28-29, no.49, pls. CCIX-X; CCXIX; CCXXI, fourth century AD. 318 DE NUCCIO-UNGARO 2002. 319 HANSEN, S. L., The embellishment of Late-antique domus in Ostia and Rome, in Patron and Pavements in Late Antiquity, S. Isager-B. Poulsen (ed.), Odense 1997, pp.115-116. 111

ROOM 7

Cat. No. 9 (a). Room 7. Polycrhome Mosaic with an orthogonal composition.320

320 Photo by author. 112

Cat. No. 9 (b). Room 7. Detail of the mosaic’s pattern.321

321 After Guida, p.58, fig.71. 113

5.3. Villa at Piazza Armerina. Mosaic of Room 25.322

Site Reference: Room 7, perhaps a triclinium.323

Dimensions: ca. m.6.50 x 8 m. The mosaic was damaged by a cut that runs north-south thorough the room when a modern aqueduct was constructed near the villa’s site in 1964. Thus, the length of its eastern side was reconstructed based on the mosaic’s design. Tesserae Count: 38 per dm on the borders; 58 on the central panel containing the stars.

Material: polychrome limestone tesserae (cm. 1-1.7).

Date: first half of third century AD.324

322 Photo by author. For this floor see CARANDINI-RICCI-DE VOS 1982. 323 For a detailed description of this floor mosaic see the contribution by CARDOSA 1999, in Atti del V Colloquio per lo Studio e la Conservazione del Mosaico Antico, AISCOM, pp.390-393. 324 M. Cardosa rejects the date of the fourth century AD proposed by Foti and dates the mosaic to the third century AD based on the stratigraphy and technical characteristics of the pavement. The size of the tesserae is similar to that of the Nereid Mosaic, which is also dated to the third century. See, CARDOSA 1999, in Atti del V Colloquio per lo Studio e la Conservazione del Mosaico Antico, AISCOM, p. 393; FOTI G., “L’attività archeologica nella Calabria,” in Atti del V Congresso di Studi sulla Magna Grecia (Taranto 1965), Napoli 1966, p.226. 114

Description: The frame of the mosaic is rendered with an isodome wall motif followed by a double guilloche (Cat. No. 9 a-b). The central field features an orthogonal composition formed by three pairs of eight pointed stars, which resulted from two interlaced squares of simple guilloches. The stars contain an octagon outlined in white tesserae, which contains a polychrome wreath of undulating band followed by two white concentric circles separated by a black circle.325 The innermost white circle contains a unitary rosette of four non-contiguous peltae with volutes unattached to the central floret.326 The cruciform spaces in between the stars are filled with a square that contains a rhomboidal motif. The square is framed by a broken meander and surrounded by four lozenges. Each star contains a wavy circle in which several vegetal motifs are inserted.

Observations: The motif of eight pointed stars formed by two interlacing squares rendered with a simple guilloche is very common all over the Mediterranean, Central Europe and Africa. The all-over pattern of this mosaic closely resembles the African prototypes. It appeared around second century AD and was employed with numerous variations for centuries across the Empire. A similar composition to that of Room 7 is found in Room 25 at the Villa in Piazza Armerina, which is dated to mid-fourth century AD (fig. 5.3).327 The same pattern is also found in other fourth century residences of Siciliy.328

325 Décor II, p.38. 326 Décor II, p.256 a. 327 BAUM 2003, pl. 22. 328 VITALE, E., “Iconografia e motivi geometrici di alcuni mosaici tardoantichi della Sicilia,” in Atti del IX Colloquio per lo Studio e la Conservazione del Mosaico Antico, 2004, p. 482-ff. 115

ROOM 60

Cat. No. 10 (a).Western register. Cat. No. 10 (b). Central register.

Site Reference: Room 60 is located near the monumental Latrina 40. In the third century phase of the villa this room was united with Room 58. In the fourth century a wall was built to separate these two rooms. Perhaps Room 60 was a waiting room connected to the Latrina 40.

Dimensions: 4 m. x 5 m. ca.

Material: limestone small tesserae (1 cm.).

Date: second century AD.329

Description: This mosaic is divided in two registers with two different geometric compositions rendered in black tesserae on a white background. The first register (Cat. No. 10 a) features a simple geometric design formed by cubes outlined in black tesserae on a white background. Each side of the cubes contains a small black square. The second register (Cat. No. 10 b) has a decoration with interlaced circles forming hexagons, which contain a five-pointed star outlined in white tesserae and filled with a black circle.

Observations: The division of the mosaic carpet in registers is typical of the second century AD and continues to be used throughout the late imperial period. The first register (Cat. No. 10 a)

329 Guida, p.67. 116

seems to be a local variation of a very widespread design. A variation of the interlaced circles motif (Cat. No. 10 b) is found at El Jem, Tunisia.330 The inscribed detail of the hexagon formed by three hourglasses creating the effect of six-pointed stars is quite diffused all over the Empire and can be found, for instance, in Pompei.331 From the time of Augustus onward in Rome and central Italy the black and white mosaics became very common as economical substitutes for polychrome designs. They involved less work in preparing the tesserae and in arranging the designs. They made it possible to cover large areas quickly and inexpensively.

330 Décor, I, pl.237, c; FOUCHER, Thysdrus 1961, pl. H, b. 331 BLAKE 1930, pl.33. 117

ROOM M

No picture is provided for this mosaic.

Site Reference: Room M.

Dimensions: ca. 1.5 m x. 0.70 m.

Material: limestone.

Date first half of third century AD.332

Description Mosaic with green and white large tesserae (cm.1.5) decorated with a motif of peltae.

Observations: For the motif of the peltae see the observations made in regards to Corridor 19, Cat. No.1 (b).

332 Guida, pp. 75, 77. 118

ROOM N

Cat. No. 11. Room N. Mosaic with black and white motif.333

Site Reference: Room N

Dimensions: Only an irregular shape portion of the pavement is extant.

Material: limestone. White, black and green large tesserae (1.5 cm).

Date: first half of third century AD.334

Description: This mosaic has been damaged by the crossing of the modern road SS 106. It features a motif of opposed bells around flowers composed of spindles.

Observations: The patterns of bells is found at Ostia in different variations on several floors.335 It is possible that this design was adopted at Palazzi di Casignana because of the influence of mosaicists from central Italy.

333 After Guida, p.75, fig.101. 334 Guida, p.77. 335 BECATTI 1961, pl.97, n.359; pl.54, n. 326; pl.99, n.408. 119

ROOM O

Cat. No. 12. Room O. Mosaic with motif of flowers.336

Site Reference: Room O.

Dimensions: The mosaic was severely damaged by the construction of the modern road SS 106. Only a small portion of the floor remains in situ.

Material: Limestone squared tesserae (2 x 2 cm.).337

Date: First half of third century AD.338

Description: This motif is composed of white unitary rosettes formed by 4 contiguous elements: 4 petals, as trifid lotus, around a green circle. Each petal is inscribed with a green small square placed vertically. The rosettes are adjacent and form green concave squares in between (Cat. No. 12).339

Observations: this pattern is found in Ostia, Italy, 340 and in Rome inside the Praetoria.341 It is much diffused throughout the history of Roman mosaics. According to S. Gozlan it is rare in Africa since only three examples have been found beside the one at Acholla.342 An elaborated variation of this motif is found in the Villa at Piazza Armerina.343

336 After Guida, p.82, fig.112. 337 Guida, p. 35, fig. 25, and p. 101. 338 Ibidem. 339 Décor, I, pl. 237, b; idem, II, pl. 275d, 340 BECATTI 1961, n. 65, p. 44, pl. 50. The mosaic is dated to the second century AD. 341 VINCENTI, V., “Mosaici inediti dei Castra Praetoria. Cenni Preliminari”, in Atti del IX Colloquio dell’Associazione Italiana per lo Studio e la Conservazione del Mosaico (AISCOM), 2004, p.268, fig.7. 342 GOZLAN S., La Maison du Triomphe de Neptune à Acholla (Botria, Tunisie), vol. I, Le Mosaïques, Institut National d’Archéologie et d’Art de Tunis, (École Française de Rome) 1992, pp.24-25, n.2. 343BAUM 2003, pl. 49, n. 1; pl. 50 n.1. 120

ROOM Q

Cat. No. 13. Room Q. Mosaic with drunken Dionysus and Satyr.344

Site Reference: Room Q, perhaps a triclinium.

Dimensions: ca. 6 m. x 3 m.

Description: The mosaic represents a drunken Dionysus (god of wine) holding a thyrsus and pouring wine into an amphora (Cat. No. 13).345 The god has long hair and is crowned with a wreath of evergreen ivy. This is a perennial that symbolizes the god’s triumph over death.346 Dionysus is supported by a Satyr who stands behind him. Near the amphora there is a stalk of vine growing from the ground. The picture is oriented west, towards Room O, because the main entrance to Room Q was located at the west before Room R was added to this part of the villa

344 After Guida, p.83, fig.116. 345 Some literary sources who describe Dionysus as the god of wine: Oppian, Cynegetica 4.230: “Now was it fated that a land, which before was wild, should cultivate the vine at the instance of Dionysos who delivers from sorrow.” , Bacchae 770: “Receive this god [Dionysos] ... I hear, that he gives to mortals the vine that puts an end to grief.”; Euripides, Bacchae 705: “Another [of the Bakkhai] let her thyrsos strike the ground, and there the god sent forth a fountain of wine.” 346 HOUSER, C., Dionysos and His Circle, 1979. 121 during a later phase of renovation.347 The figures of Dionysus and the Satyr are drawn with the characteristics of the silhouette style that was widespread from the Severan period and widely represented on black and white figural mosaics at Ostia and Rome.348

Material: large limestone green tesserae and white marble tesserae.349

Date: first half of the third century AD.350

Discussion and Comparisons: A close iconographic resemblance of this floor can be seen in a polychrome mosaic from Turkey, which is dated to the fourth c. AD (fig. 5.4). There the drunken god walks supported by a Satyr and spills his cup of wine in the mouth of a panther cub. 5.4. Museum Collection: Antakya Turkey Context: Antioch, House of the Drunken Dionysus Date: 4thc.AD.351

The choice of representing a drunk Dionysus inside Room Q may be linked to a probable production of wine on the estate of the villa at Palazzi di Casignana. In fact, a large amount of wine amphorae Keay 52 manufactured in Bruttii were found in situ. On the opposite side of the villa lies Room B, which features a mosaic with the Four Seasons. These two themes, Dionysus and the Seasons, may have been chosen by the villa’s owner to evoke fertility and prosperity on his land. The god of wine may also simply signify the pleasure of drinking and conviviality during the symposium that may have taken place in Room Q.

347 Guida, p.82. 348 CLARKE 1979, pp. 87 ff. 349 Guida, p.101. 350 Ibidem. 351 from http://www.theoi.com/Olympios/DionysosGod.html#Viticulture

122

In fact the theme of the wine, conviviality, happiness and joy of living is appropriate for rooms in which banquets were held. The extensive use inside homes of figurative tapestry (see for instance that represented on fig.3) and all kinds of luxury items reproducing scenes or themes of the Dionysian cult likely contributed to the long history of the Dionysian iconography.352

5.5. Panel with the Triumph of Dionysus, 4th century; Said to be from Panopolis (Akhmim), Egypt. Undyed linen and purple wool, tapestry weave.353

D. Parrish has pointed out that the diffusion of the Dionysian motifs throughout Late Antiquity is partly due also to the knowledge of the Dionysiaca, an epic tale of the god Dionysus written by Nonnos between the fourth and fifth century, later than the construction of Room Q. The Dionysiac figures and themes occur in exceptional profusion on northern African mosaics, not distant from Southern Italy, from the early second century AD to the late fourth. These belong to a general artistic koine from the onwards and have little religious significance as it is the case for the Nereids and sea-monsters of the marine mosaics.354 Dionysus is also the god of death. According to K. Dunbabin, when Dionysus is associated to the Seasons on sarcophagi dated to the Severan period onward the god may represent a belief if the after-life and the seasons would illustrate the eternal regeneration of life.355 If the Villa at Palazzi di Casignana hosted Christian cult and the burials around the villa had some association with it, the Dionysiac subject would still be considered appropriate for

352 BALDINI LIPPOLIS 2001, p.75; PARRISH 1984; Nonnos, Dyon., 9. 353 After http://www.metmuseum.org/toah/ho/05/afe/ho_90.5.873.htm 354 DUNBABIN 1978, p.173 with numerous examples of mosaics representing the Dionysiac theme. 355 Ibidem, p.186; HANFMANN, G.M.A., The Season Sarcophagus, I, 1951, p.236. 123 embellishing the floor of one of the main reception rooms of the building. In fact, as C. Houser states:

“Roman forms of Dionysus passed into Early Christian imagery with barely a change; the transition was easy and logical because of the parallels between the life of Dionysus and Christ: both were the son of the father god (God the Father) and a mortal mother, both dwelt among mortals, and gave their followers the promise of resurrection after death. Wine is part of the ritual of both and the Eucharist Host and the legendary consumption of raw flesh (omophagia) may have similar symbolism.”356

Yet, it is more likely that the Dionysus of Room Q because he of his drunkness is associated with the probable production of wine on the villa’s estate rather than with some Christian symbolism.

ROOM R

Cat. No. 14 (a). Mosaic of Room R.357

356 HOUSER 1979, pp.15-16. 357 After Guida, p.83, fig.115. 124

Cat. No. 14 (b). Emblema of the mosaic in Room R.358

Site Reference: Room R. It is a small circular open lodge with three columns.

Dimensions: ca. 3.5 m. of diameter.

Material: limestone large tesserae (1.5 cm.) on the mosaic field. Small tesserae on the emblema.

Date: Second half of the third century AD.359

Description: Round polychrome mosaic framed by a vegetal motif in black and green tesserae (Cat. No. 14 a). The emblema360 represents a female face, perhaps a personification (Cat. No. 14 b). It is outlined in black tesserae, with the flesh in white and the details in green and red tesserae. The face is quite damaged but the remaining parts display good skill of execution.

Observations: The female figure on the emblema is not identifiable. For a probable function of Room R see discussion in Chapter 4 of this study.

358 After Guida, p.82, fig.113. 359 Guida, p.82. 360 An emblema is a detailed, and often quite small, part of a mosaic which was usually prepared in a workshop beforehand. It was then inserted into a larger background area of less complicated design that was easier to create on site. 125

ROOM I

Cat. No. 15 (a). Room I. Mosaic of northern arm.361

Site Reference: Room I, northern arm of the cruciform plan.

Dimensions: Only a small portion of the floor remains in situ. The mosaic originally covered an area that measured circa m.4.50 x 3.70.

Description: Six-pointed star of white tesserae around a black hexagon inscribed in a white hexagon forming a composition of honeycomb web. Lateral band: square of swastika meander in black tesserae on a white background (Cat.No. 15 a).

Material: black and white limestone tesserae.

Date: first half of fourth century AD.362

Discussion and Comparisons: the six-pointed star motif dates to the first century BC and lasted throughout the history of Roman mosaics. There are many mosaics featuring this motif and several variants all over Italy and the Empire. A close parallel to our mosaic, but with reverse colors, can be found on a doorway at the House of Livia on the Palatine Hill in Rome, which is dated to the first century AD. There, the star is rendered in black tesserae standing on a white ground.363The same motif recurs also at Aquileia.364 Another close parallel to our mosaic is found in Piazza Armerina.365

361 After Guida, p.85, fig.121. 362 Guida, p.103. 363 MORRICONE MATINI 1967, p.56. 364 See Decor, I, p. 328 e. (Fondo Cossar). 365 BAUM 2003, pls. 18-19, nn.1-6. 126

Cat. No. 15 (b). Room I: central part of the floor mosaic.366

366 After Guida, p. 84, fig.119. 127

Cat. No. 15 (c). Detail of the composition.367

Site Reference: Room I: central part of the floor (east-west orientation).

Dimensions: not determined.

Description: A square is inserted diagonally inside a larger square. Both squares are framed by a guilloche (Cat. No.15 b). The smaller square contains a rosette (Cat. No. 15 c). This pattern is combined with an orthogonal composition of swastika meander created with a vegetal motif. These are recessed and reversed to form a T design (Cat. No. 15 b). The corner of the mosaic that decorates the central part of Room I features a vegetal motif, an acanthus scroll, rendered in green and red tesserae on a white background (Cat. No. 15 d).

Material: limestone, polychrome tesserae.

Date: first half of the fourth century AD.368

367 Afger Guida, p. 85, fig.120. 368 Guida, p.103. 128

Discussion and Comparisons: The motif of a square inserted into a larger square is found frequently all over the Empire in different periods. The smaller square containing a rosette (Cat. No. 15 b) finds a close parallel in two mosaics at Piazza Armerina.369 The swastika meander is used to decorate a mosaic in Gaul and is frequently attested in both the western and eastern Empire.370 The use of vegetal and floral designs, such as the one depicted on the corners of Room I, is widespread all over the Empire. Plants, leaves and flowers are often rendered in a very naturalistic style. These vegetal motifs were used to fill parts of the floor, to decorate its frame, but also to cover the entire mosaic’s surface. Sometimes geometric patterns are ‘vegetalised’ by outlining the geometric shapes with fine tendrils or rows of petals.371 On northern African mosaics, branches of fruit and flowers may be depicted randomly over the floor.372 The acanthus scrool at the corner of Room I is one of the most common vegetal motif adopted by Roman mosaicists. It originated as a border motif and continued to be used as such for several centuries. Perhaps it derived from vase-painting and architectural decoration. The mosaic workshops created distinct variations following a common and standard design. As a result, there is a wide variety of this vegetal motif all over the Roman world.373 It is found in several geographical and chonological contexts, such as Timgad (Algeria),374 Nebo (Palestine),375 and Italica (Spain).376

369 BAUM 2003, pl. BB, 1 (Room 31), and 2 (Room 38a); pl. 21,2; pl. 71,3.. 370 Recueil IV, 1, pl.184, n.460. For the meander motif see GUIMIER-SORBETS, Le méander à pannetons de clef dans la mosaïque romaine, in Mosai , Paris 1983. Hommages à H. Stern, pp.195-213. 371 For a treatment on the floral and vegetal designs see DUNBABIN 1999, p. 296 ff. 372 Ibidem. 373 Ibidem for a description of several examples. 374 Ibidem, fig.20. 375 Ibidem, p.198, fig.210. 376 Ibidem, p.151, fig.156. 129

Cat. No. 15 (d). Detail of the mosaic’s corner with vegetal motif.

130

ROOM B

Cat. No. 16 (a). Room B. Mosaic with the Four Seasons.377

377 After Guida, p.86, fig.123. 131

Cat. No. 16 (b). Lateral frame of mosaic in Room B.378

Site Reference: Room B, Triclinium?

Dimensions: m.8.20 x 6.70.

Description: polychrome mosaic featuring a geometric motif of interlaced meanders rendered in green, black, white and red tesserae. Four squares are drawn at the corners of the room framed by a laurel motif in green and white color. Two of them contain an inset or emblema (circa 8-9 per square cm.) with a female bust (Cat. No. 16 a). These images represent the personification of Spring and Winter and they are placed at the NE corner (Spring) and NW corner (Winter) of the room. The other two Seasons (Summer and Autumn) are no longer extant. They were also inserted in squares at each corner of the southern side of Room B. All four Seasons face East towards the room’s entrance, which provides the optimal viewpoint. The lateral frame of Room B (Cat. No. 16 b) is decorated with a polychrome orthogonal pattern of intersecting circles, forming saltires of quasi-tangent dentilled spindles. The circles are inserted with concave squares.

378 After Guida, p.86, fig.122. 132

Cat. No. 16 (c). Room B: Spring.379

The image of Spring (fig. Cat. No. 16 c) is crowned with flowers and leaves alternatively rendered in red and green. The face is rounded, a shape that suggests prosperity. The woman turns slightly to her right and looks towards the spectator. Locks of hair are outlined in black tesserae. Lines of different shades inside the face reveal the intention of the mosaicist to give volume and realism to the figure. Yet, the result is little messy. Big eyes outlined in black create a rigid expression with little depth; the pupil and are both black and treated with scant realism. The nose is straight and short. It creates a black and rigid shadow underneath. The lips are small and red outlined with two black lines. The chin is rounded and creates a shade on the neck rendered with dark pink lines. A rich garment wraps around the woman’s neck. On the right side of the figure, a stalk with green leaves and three red flowers on the top rises from the bottom. The rendering is a little rigid. This female bust is a work of good quality but shows a different manufacture from the image of Winter, which is rendered with greater skills and more refined style. It is possible that the two emblemata are the work of two different officinae. Female representations of the seasons on mosaics and in Roman art in general have characteristic features.380 They are usually placed on the diagonals or corners of the floor,

379 After Guida, p.86, fig.124. 380 For a thorough study of the Seasons in both literary and figurative sources see, HANFMANN, G., The Seasons Sarcophagus in Dumbarton Oaks, Cambridge-Mass. 1951, II, n.176, This works also includes a catalogue of all the monuments, divided in different categories, which feature the subject of the Seasons. 133

following a scheme that originated in the Hellenistic age. Occasionally, they have wings but the most distinctive attribute, on several mosaics across the Empire, is represented by roses. On this mosaic roses appear on the head crown and on top of the stalk alongside the figure. They celebrate the blooming of nature during Spring and are depicted in a large variety of forms as, for instance, a garland in the woman’s hair, a bouquet in a basket, or spread around the female figure.381 Another characteristic trait of Spring is an elaborate, richly decorated garment. On this mosaic the garment is wrapped around the woman’s neck. It may refer to the elegant costume that women wore during the Rosalia, or the festival of roses, which was celebrated in the month of May in many parts of the Empire. 382

Material: Polychrome limestone tesserae.

Date: first half of fourth century AD.383

Observations: The orthogonal pattern of the lateral band (Cat. No. 16 b) finds a close parallel at El Jem, (Tunisia),384 Acholla (Tunisia),385and Piazza Armerina (Sicily).386 The iconography of the two Seasons inside Room B at Palazzi di Casignana finds a close parallel in one mosaic at Sousse (Tunisia), which is dated to the end of the second century AD. According to D. Parrish, the figure of Autumn on the mosaic at Sousse, resembles some painted portraits of the time of Septimius Severus because of the oval face with large eyes, heavy eyelids, arching brows, and a shadow under the chin.387 It is possible that the mosaicist of Room B at Palazzi di Casignana used an iconography that was in use already during the Severan period. The long adoption of the Seasons’ iconography makes it difficult to date these mosaics with precision. The representation of the Seasons on floor mosaics is widely spread all over the Roman Empire from the first half of the second to the sixth century AD. In Italy more than 50 examples are known. In Bruttii, a floor mosaic representing the Four Seasons was found at Vibo Valentia inside a third century aristocratic house (fig. 5.6). Seasons may be represented with female or male personifications, often accompanied by images of Earth, the Ocean and other cosmological deities. At Casignana, it is very significant that within the Eastern area of the villa two mosaics refer to propitious subjects: the northern side contains Room B, which is decorated with the Four Seasons mosaic; the southern part includes Room Q that features a Dionysiac mosaic. The introduction of these two themes in the decorative plan of the villa was clearly intentional. The Seasons signify the perennial continuity of time and the eternal revival of nature. Since the Villa at Casignana is located on a fertile area in front of the sea, very likely they allegorically referred to the fertility of the land, and the consequent prosperity of the villa. On the other hand, Dionysus of Room Q may refer to a possible production of wine on the villa’s estate. But the god of wine also symbolizes endless regeneration. As G. Canuti points out, from the early imperial period onward, the frequent recurrence inside the wealthy residences of the Roman nobility of

381 See PARRISH 1984, p.34. 382 Even the personification of May wears an elaborate and elegant garment on a mosaic from Carthage. See PARRISH 1984, p.34 with notes. STERN 1953, pp. 249-252. 383 Guida, p.103. 384 FOUCHER 1961, pl. H, b. 385 Antiquités Africaines, CNRS, Paris, 1968, p.105, fig. 5. 386 BAUM 2003, I, pl.T, 4: Room 39c. 387 PARRISH 1984 , pl. 81a. 134

such iconographies, as well as those representing the zodiacal signs, the months, sea, rivers, sun, Aion and Ocean, is very significant. It reflects the assimilation of imperial propagandistic themes within the private, domestic sphere. In the manner of the Hellenistic monarchs, from the time of Augustus, Roman emperors promoted the idea of renovation, prosperity, peace and endurance of the Roman Empire in many public monuments and also on coinage. The image of the Seasons, for instance, appears on reliefs of the Trajan Arch at Beneventum, and later on the Arch of Septimius Severus and that of Constantine in Rome with the purpose of celebrating the return of a golden age under the auspices of a new . At the same time, ‘Dionysus’ was the epithet used by several emperors such as Trajan, Hadrian and Antoninus Pius who wanted to appear as the restorers of the Empire.388 A parallel between the fortitude of the Empire and the endurance of the Cosmos was firmly established in figurative art. Under the order, peace and stability that the emperor provided to his subjects, Earth could enjoy abundance, prosperity and the renewal generated by propitious seasons. This powerful idea was brought inside the domestic sphere of the wealthy Romans and translated into symbolic images such as those of the Seasons. These figures delivered a message of perennial good fortune inside the house, which was the place for the nourishment, comfort, wealth and endurance of the familia. As the emperor provided prosperity to his subjects, so the aristocratic dominus assured wealth to his family by living in total harmony with the progression of time, the development of nature, and the transcendent forces.389 The self-celebration of the dominus at Casignana is addressed also by means of a mythological iconography such as that of the Seasons and Dionysus on mosaic floors. From the Antonine age, the faith in providential deities and the ideology or renewal, fecundity, and wealth lasted well into the late Empire. of Costantius I (337-350) and Constantius II (337-361 AD), reproduce images of the Seasons with the inscription feli[icium] or felix temp[orum reparatio.

388 CANUTI, G., “Considerazioni sulle origini ideologiche e culturali dei pavimenti a tematica cosmologica nelle domus fra il II d.C. e tarda antichità, in Atti dell’XI Colloquio dell’Associazione Italiana per lo Studio e la Conservazione del Mosaico (Ancona, 16-29 febbraio 2005), 2006, pp. 57-64. 389 Ibidem. 135

5.6. A mosaic with Seasons in Vibo Valentia (Calabria), Italy.390

390 Photo of author. 136

Cat. No. 17. Room B: Winter.391

This emblema is inserted inside a square framed by a laurel motif. It is placed above the image of Spring, in the upper right corner of Room B. The female bust represents the personification of Winter as recognizable by the green hooded mantle (referring to the cold weather), and the head garland made with green olive leaves and their dark- red fruits, which refer to the olive harvest and its season.392 The face of Winter turned slightly to her right and displays a fixed glance rendered in a similar manner to that of Spring. The slimmer face of Winter suggests the old age of the season that comes latest in the year. In one mosaic from La Chebba in Africa, all four personifications of the seasons show a progression in age to make an analogy between human age and the passage of the year. Since two emblemata of the Seasons are lost, it is not possible to say if a similar progression occurred on the mosaic of Room B at Casignana. A close parallel to this mosaic is found in a mosaic at Sousse, the date of which is debated. 393On stylistic and archaeological ground the mosaic of Room B dates to the beginning of fourth century AD.

391 After Guida, p.87, fig.125. 392 PARRISH 1984, p.32. 393 D. Parrish dates it to the Severan times, whereas Foucher to the end of third century AD. See PARRISH 1984, p.233; FOUCHER 1960, p.20, No.57.042. 137

CONCLUSIONS

The owner of the Villa at Casignana explicitly desired mosaics that created an atmosphere of wealth and opulence in order to convey a message of high social status and economic power. The floor mosaics so far uncovered display both a traditional style and a more modern taste that was typical of the later Roman centuries. The simple geometric mosaics that adorn several rooms of the Western Baths, such as that of Room 12 with tridimensional cubes, or Room 24 with elongated hexagons belong to the old and common tradition of black and white mosaics found all over Italy and the Empire since the late Republican and early imperial period. The adoption of these patterns for several centuries provides a good examble of the mosaic art’s conservatism. The date of mosaics adorned with the black and white motifs can be easily mistaken if based exclusively on their style without considering the archaeological and stratigraphic data. On the other hand, the polychrome and elaborate geometric polychrome compositions, such as those of Room 7 and Room 36, belong to a more recent fashion influenced by African mosaicists and their workshops.394 These mosaics reflect skills and a well balanced use of colored tesserae accurately oultlining each design. The use of colors is not exaggerated but limited to four or five colors. Few floral designs are mostly confined to the corners of the floors, as the acanthus scroll depicted in the corner of Room I. The style of mosaics in Room 7 and Room 36 originated in the Proconsular province (modern Tunisia) and developed in Italy and the western provinces of the Mediterranean between the third and early fourth century AD. It was a style characterized by complex geometric patterns rendered with tesserae of multiple colors, use of vegetal elements such as acanthus scrolls and laurel wreaths, and figural representations executed with great realism.395 There is evidence of frequent contacts between southern Italy and Africa as indicated by a large importation of African sigillata and the circulation of patterns or the movement of craftsmen from one area to the other of the Western Empire. This created within the mosaic art a koine, or a common decorative language, which often blended with the local tradition of each region. As a consequence, some third and fourth century AD mosaics of Bruttii found at Casignana, in Vibo Valentia and Trainiti/Briatico396 share a resemblance with the African mosaics. It is possible that the local ateliers summoned a few craftsmen from Africa or were simply influenced by African copy-books that circulated among mosaicists.397 Other media such

394 The most outstanding example of the African style is represented by the mosaic floors inside the villa at Piazza Armerina, dated to the time of Constantine. They were created by workshops that either were brought to Sicily for the commission of the villa’s owners, or transferred permanently to the island. At the Villa of Desenzano, for instance, alongside the mosaic representing Erotes fishing in a sea, a realistic image which finds parallels in African mosaics, there are others that are composed in small sections according to a traditional Italian fashion. 395 Figural mosaics of the African style typically draw their repertory from , amphitheatre, circus and rural life of large estates, but they also represent mythological scenes. Designs are created with big polychrome figures that are larger than in earlier periods, arranged in registers or placed freely over the pavement, which is treated as a two-dimensional surface. The landscape or natural features are often rendered with partial perspective. Instead of thin lines, the patterns are outlined by broad bands of guilloche, wavy ribbons, and broken meaner. On African mosaics and their style see LING 1998, pp. 77-97; DUNBABIN 1978, and 1999, pp.101-129. 396 For the mosaics in Vibo Valentia and Briatico/Trainiti, see FAEDO 1994, pp.601-606 and 614-615. For a survey on the existence of a “southern Italian style” see DUNBABIN, K.M.D., “The San Giovanni Mosaic in the Context of Late Roman Mosaic Production in Italy,” from the: Lo Scavo di S. Giovanni di Ruoti ed il period tardoantico in Basilicata. Atti della Tavola Rotonda , Roma, 4 luglio 1981. 397 On the circulation of cartons and craftsmen see DUNBABIN 1999, pp.269-278. 138

as paintings, architectural decorations, stucco, sculptures, metal works, pottery and textiles contributed to a common decorative repertoire shared by the Roman mosaicists. At Casignana, the figural mosaics that represent the Four Nereids (Room 9), a drunken Dionysus (Room Q), and the Four Seasons (Room B) all used an iconography that was widespread for centuries throughout the Western Empire. These scenes were laid out in accordance with the size and shape, orientation, and likely the function of each room. The minimal character adopted in the Nereid mosaic of Room 9, which in different contexts is treated with greater complexity, suggests that the mosaicists who worked at the villa might have not been familiar with elaborate figural representations. The interpretation of the Nereids’ myth is rather simple, with the figures of the four deities outlined in white tesserae that float on the unitary surface of the sea rendered in green tesserae. The handling of the figures and the whole arrangement is well equilibrated creating a refined effect. Even the Dionysiac theme at Casignana is represented in a simple arrangement. The room is rather small, so using a more composite representation of the subject would have produced a chaotic effect. Only two figures, the drunken god and the satyr, are drawn in linear and graphic style. There are no decorative details inside each figure, and a very plain and simple scheme dominates the floor. The third figural mosaic at the Villa, that representing the Four Seasons with the surviving image of Spring and Winter shows a schematic realism and a uncertain use of polychrome tesserae. Subtle differences can be seen on the same floor, which might be attributed to the work of several mosaicists who had different skills, or to slightly different chronological phases of the floor’s execution. Overall, the mosaics at Casignana show a particular simplicity and reflect the general koine of the western Mediterranean. They do not depart from the style and iconography that developed during the latest centuries of the Empire. The polychrome geometric patterns parallel the North Africa repertory. Further studies of the villa’s mosaics, their style and the way the tesserae were laid out might help to distinguish several teams of craftsmen who worked at the villa during different phases of restoration. Some craftsmen may have worked at decorating the reception rooms, while others worked on the less prestigious spaces. Another specialized team of mosaicsist may have worked exclusively on the vaults of those rooms that were adorned with glittering mosaics, such as Room 12, Room 25, and Room I. The mosaics at Casignana all reflect the wealth of a prosperous area and a sophisticated commission. In Calabria (Bruttii) more mosaics need to be studied in their original context and then published in order to determine the distinctiveness of the workshops that operated in the region during the Roman times. This would also help to identify the ateliers that worked at Casignana and their relationships with other centers of the Roman world, especially with Africa, Sicily, and Central Italy.

139

The use of marble inside the Villa at Palazzi di Casignana

5.7.Room 9: opus sectile floor.398

Marble is used abundantly inside the Villa at Palazzi di Casignana for wall veneers, thresholds, columns and an elegant opus sectile floor (Fig. 5.7). The wealthiest Romans all over the Empire embellished their residences with this expansive material even during Late Antiquity.399 The walls covered with marble must have provided a precious background for sculptures, textiles, painted pictures and portraits that likely furnished the villa. The variety of marble used at the villa, such as the Pentelic and cipollino from Greece, the pavonazzetto from Phrygia in Asia Minor, and the red porphyry from Egypt, point to a flourishing marble trade between Bruttii and distant regions of the Empire. These areas have also produced numerous examples of opus sectile floors.400 The large quantity of marbles that were imported to Rome and other centers of the Empire required a complex and sophisticated organization not only for the stone extraction but also for its transportation. During the Late Republic the quarries were privately owned but under the imperial period a large number became property of the emperor either by conquest, purchase or inheritance. Many quarries remained the property of very wealthy Romans but their production was inferior to that of the quarries owned by the emperors. The latter produced a large quantity of marble,

398 Photo by author. 399 See, for instance, the villas of Aquitania, in BALMELLE 2001, p.222. 400 DUNBABIN 1999, p.261. 140

which was mainly used to embellish public buildings or sold to private patrons for the decoration of their villas and residences. We know very little about the private quarries of the early Empire. The trade of marble certainly decreased during the third century economic crisis that affected the Empire.401 At this time, the marble trade dropped to such an extent that in the fourth century the emperors had to encourage private parties to exploit their marble quarries by promulgating laws that established that no taxes had to be paid for extracting marble. The first of such laws was issued by Constantine in 320.402 It was addressed to the fiscal procurator of Africa and established that private owners of quarries had the total freedom to extract marble without being taxed. As a result, private quarries became very productive and competitive with those owned by the State. In fact, taxation on the extracted marble from private quarries was reinstated again.403 It is possible that the marble used inside the Villa at Palazzi di Casignana was purchased, but it is not excluded that it was extracted in quarries owned by the villa’s dominus. It is precisely during the period when the villa was renovated (between the third and fourth century AD) that private owners were encouraged to reopen their quarries in order to extract marble, put it on the market and meet the high demand of wealthy patrons all over the Roman world. Whatever was the case, the array of colored marbles used at Palazzi di Casignana suggests that the villa’s dominus operated in a commercial network that could easily furnish this precious stone in accordance to his desire and taste. The trend among the late Roman aristocracy to acquire precious and rare marble is confirmed by a poem written by Sidonius Apollinaris in the fifth century AD. Sidonius praises the marvelous marbles from Gaul that decorate the private baths at Burgus of his friend Pontius Leontius, whom he describes as the richest and most noble man of Aquitania. The poet compares them with the marbles from other regions of the Empire: Ipsa autem quantis, quibus aut sunt fulta columnis! Cedat puniceo pretiosus livor in antro Synnados et Nomadum qui portat eburnea saxa collis et herbosis quae vernant Marmora venis; candentem iam nolo Paron iam nolo Caryston; vilior est rubro quae pendet purpura saxo (How many columns these magnificent baths are decorated with! The precious scarlet color that adorns the den of Synnados, the hills of Numidia which contain ivory rocks and marble with veins of variegated green plants should yield to the columns of Burgus. I do not want either the shining marble of Paros, or that of Carystus. The purple hanging from the coral rocks cannot match the magnificence of the marble of Burgus).404 The exotic marbles imported from the East offered an astonishing array of colours ranging from whites, blacks, grays, reds, yellows, blues and greens. Their extraction, the organizational challenges, logistical difficulties and sheer expense of their trasportantion westward, carried a powerful ideological meaning. On any visitor to walk through the Villa at Palazzi di Casignana and admire its marble veneers certainly produces an awe-inspiring effect. These rare and costly stones expressed limitless wealth and power reflecting the elevated status of the villa’s patron.

401 GNOLI 1971. 402 Codex Theodosianus, 10, 19, 1, see GNOLI 1971, pp.28-29. 403 GNOLI 1971, p.29; Codex Theodosianus, 10, 19,10. 404 Sidonius, Carmen 22; GNOLI 1971, p.39. 141

CHAPTER 6: Late Roman Villas in the Territory of Locri: Four Remarkable Examples

In order to examine the economic conditions of Bruttii in the fourth century AD when the Villa at Palazzi di Casignana was renovated, and to set the monumental phase of this villa more clearly in its historical context, it is essential to look at the other villas in the territory of Locri. Other rich residences were built before, after or during the same period when the Villa at Palazzi di Casignana expanded to attain its greatest splendor. Although the four villas that will be described below were not as huge and lavish as the Villa at Palazzi di Casignana, they demonstrate that during the late imperial centuries in Bruttii a dynamic Roman aristocracy lived and prospered, forcing us to reconsider the old perception that the region underwent no major development during Roman times. The archaeological evidence shows that between the third and fourth century AD the aristocracy of Bruttii was every bit as dynamic and successful as its counterparts elsewhere in the Empire. This, in turn, stands as proof that no matter where they lived, Roman aristocrats of this period led a similar life-style seeking leisure and entertainment while still managing business and lucrative enterprises. They were generally motivated by these commonly shared values and goals. The Roman elite conformed to the same social behavior everywhere: behavior that was dictated by a strong sense of belonging to a very exclusive group. In a very recent study, J. Perkins writes about the Roman imperial elite’s character in such terms: “a trans-empire group identity evolving in the early empire of persons bound together by ties of privilege, education, culture, and connections with the imperial centre and by the shared self-identity these ties constituted.”405 According to this scholar, the identity of the imperial elite was constructed by placing the classical at the center of the cultural and educational milieu. The Herodotean conception of ‘Greekness,’ which is based on the sharing of blood, speech, and way of life (, 8.144) is no longer dominant. Rather, from the early Empire onward, paideia is the driving force that gathers people of the elite from different backgrounds and origins, and which makes the elite a distinguished and superior group from the rest of society. Paideia, according to Perkins, cements “social ties among the imperial trans-national elite”406 and causes the formation of an endogamic society in the multicultural world of the Roman empire. Moreover, Perkins argues that the “elites from across the eastern empire joined with Rome in the management of its empire, and in turn received Rome’s support for their position and privileges, as long as they helped ensure peace and tranquility within their respective cities and avoided civic problems.”407 Until recently, Roman Bruttii was considered a region of the Roman world that did not generate or leave behind significant monuments, especially when compared to the same region in the Greek era.408 Today, however, the knowledge of Roman Bruttii has improved considerably as a result of the archaeological discoveries of the last 15 years, using improved scientific methods. Discovery of late imperial villas in the region has sparked a new interest in the later phase of Roman Bruttii. Information from late antique authors is being revaluated as new data and archaeological material come to light.409

405 PERKINS, J., Roman Imperial Identities in the Early Christian Era, 2009, p.5. 406 Ibidem, p.25. 407 Ibidem, p.23. 408 A critique of this approach is in ARSLAN 1998, p.397. 409 G. Sangineto catalogues 22 villas dated to the late Roman centuries in Bruttii. See SANGINETO 1994, pp.560- 575. 142

Still, very few Roman villas and sites from the region have been systematically excavated or published and a thorough study of their territories may provide more proof of the dynamism of the Romans in this region. A broader topographical exploration would offer a clearer picture of life inside and around each of these villas, the economic role of these properties, and how they were used by the elite. The villas described in this chapter represent outstanding examples of aristocratic dwellings in the region of Locri from the third to the fifth century AD. These residences are characterized by good planning, luxurious amenities, and in some instances agricultural and industrial facilities. They are (see map on fig.6.1): 1. The Villa at Marina di Gioiosa Jonica (1st c. BC- 4th c. AD); 2. The Villa of Naniglio in , not far from the seaside town of Marina di Giosa Jonica (1st –4th c. AD); 3. The Villa at Ardore (3rd-4th c. AD); 4. The Villa at Quote San Francesco (5th- 6th c.AD).

143

Quote S. Francesco

6.1. Map of Calabria. Roman sites near Locri.410

410 After Guida, p.22, fig.8. 144

1) The Villa at Marina di Gioiosa Jonica (fig.6.2).411

About 12 Km. northeast from Locri, in the town of Marina di Gioiosa Jonica, archaeologists uncovered in the 1920s a villa that had a life spanning from the first century BC to the fourth century AD. The villa during its third century AD phase featured private baths decorated with rich marble and floor mosaics.412 The rooms and tubs are still partially visible.

6.2. The theater at Marina di Gioiosa Jonica(3th c. AD).413

Remarkably, near the villa a small well-constructed Roman odeum was discovered, which had a capacity of 1200 spectators (fig.6.2). U. Kahrstedt, E. Arslan and S. Ferri maintained that this theater was part of the nearby villa as well as the baths. From its architectonic characteristics they dated this building between the third and fourth century AD.414 Today the theater stands by a modern train track, but in antiquity it was well inserted in the urban plan of the ancient Marina di Gioiosa Jonica. The theater was built on a plain. The cavea was formed by low concentric walls, which were likely covered with terracotta or stone slabs to create seats. Only 10 rows of the cavea survive, but Ferri calculated that there were originally 20 allowing the odeum to accommodate as

411 After De Franciscis 1988, fig.74. 412 KARHSTEDT, ARSLAN, and FERRI, Notizie Scavi 1926, pp. 332-335. 413 After PAOLETTI 1994, p. 513, fig.56. 414 The theater today is surrounded by modern structures. A thorough report of the theater is in FERRI, S., “Gioiosa Jonica Marina. Teatro romano e rinvenimenti rari,” in ‘Notizie degli Scavi”, 1926, pp.332-338; COSTAMAGNA, L.,- SABBIONE, C., Una città in Magna Grecia. Locri Epizefiri. Guida Archeologica, 1990, p. 302; KAHRSTEDT, U., Das wirtschaftliche Gesichte Griechenlands in der Kaiserzeit : Kleinstadt, Villa und Domane, Wiesbaden 1960, p.65; ARSLAN, E.A., “La ricerca archeologica nel Bruzio,” in Brettii, Greci e Romani. Atti V Congresso Storico Calabrese (Cosenza-Vibo Valentia-Reggio C. 1973), Roma 1983, pp. 271 ff. 145

many as 1200 spectators.415 An amphora imprinted with a figlina stamp (a brick workshop stamp) was found inside the theater; the stamp reported the name Satur(nini). This suggested that Saturninus was the likely owner of the figlina located near the villa. Another inscription found in Locri, which was part of a funerary monument dated to fourth century, contained the same name of Saturninus. The inscription, perhaps a confessio, reads Saturninus v(ir) c(larissimus), referring to an aristocrat of senatorial rank. It is possible that the inscription found in Locri commemorated the same Saturninus who owned the figlina in Marina di Gioiosa.416 It was not uncommon for Roman senators to be involved in the production of bricks on their estates, as a rewarding source of income. 417 Arslan compared the theater of Marina di Gioiosa Jonica to that of the ‘Villa of Agrippa,’ or ‘Baths of Agrippa,’ on the Tuscan island of Pianosa (fig. 6.3).418 This villa was owned by Agrippa Postumus (12 BC-14 AD), grandson of Augustus, who was exiled on the island and there murdered.419

415 There are four staircases that divide the cavea in five cuneal sections. The orchestra is at a lower level in respect to the cavea and had a diameter of 14.20 m. It is enclosed by a high wall, upon which portions of painted plaster survive featuring decorative squares framed by a red band. 416 COSTABILE 1976, p. 121. 417 The name Saturninus is found again in a Christian context at Scolacium, see COSTABILE1976, p. 121; CIL X, 102. The figlinae (industrial establishment of bricks) were usually located near clay deposits and along water channels that allowed the transportation of the products. The workshops often belonged to important persons linked with the imperial family. The bricks were marked with a stamp that provided indication of the producers. 418 ARSLAN, E.A., “La ricerca archeologica nel Bruzio,” in Brettii, Greci e Romani. Atti V Congresso Storico Calabrese (Cosenza-Vibo Valentia-Reggio C. 1973), Roma 1983, pp. 271 ff. 419 The villa was built in the first century BC, and abandoned by the first half of the first century AD. See SCHMIEDT 1972; MIELSH 1987; PANCRAZZI AND DUCCI 1996; HIGGINBOTHAM 1997; TOSI 2003; MASTRAGOSTINO, L., I tesori dell’isola di Pianosa nel Mar Tirreno, Firenze 2001; MARZANO 2007, p.693; LAURENZI , E. “Edifici di spettacolo in contesti privati: la Villa di Pianosa,” in BURRI-DELACRETAZ- MONNIER-NOBILI (ed.), Ad limina II, 2004, pp.188-189. 146

6.3. The Roman Theater at Pianosa (1st c. AD).420

Both theaters were built near baths and were small. The theater at Pianosa had five or six rows of seats and was built on a rocky cliff. Its cavea had a diameter of 13 m. and in the center features a platform with an apse (5 x 2.50 m.) decorated with polychrome marbles. The cavea and scaenae frons featured white marble decorations. In 1926, Ferri reported that several artifacts were found in the area of the villa at Marina di Gioiosa Jonica, including sculptural fragments. One was an exquisite fragment of a female statue.421 Ferri believed that they belonged to the villa, as well as the theater and the baths. Recently it has been hypothesized that the theater, the baths and the other findings uncovered by Ferri belonged to an inhabited center identified as a statio, posing a number of questions about the villa’s function and correct interpretation. Arguments for assigning the theater and other buildings to a statio are not strong enough to make them preferable to the suggestion that they indeed belonged to this villa, as was suggested at the time of their original discovery. The statio in this area – known from itineraries- that include all of these structures would have to be Subsicivo, a place near Marina di Gioiosa. Yet, the name Subsicivo, in Latin means “left out” and, as G. P. Givignano has proposed, may refer to a locality that was “left out” from the repartition of land near Locri, i.e. it should not be here in Marina di Gioiosa Jonica. 422 The two surviving Roman itineraria offer two different locations within the same area making the identification of Subsicivo highly hypothetical.

420 After http://piccard.esil.univmed.fr/venus/missions/pianosa/pianosa_intro1.html 421 FERRI, S., in Notizie Scavi, 1926, pp. 337 ff., fig.5. For the fragment of the female statue see FAEDO, L. 1998, p.629. 422GIVIGLIANO 1994, p.323, note 386. 147

Apparently, the that the theater at Marina di Gioiosa Jonica belonged to a nearby statio has been hypothesized on the grounds that it would be truly exceptional to find a Roma villa with its own theater in Bruttii. But the discovery of the Villa at Palazzi di Casignana proves that wealthy residences equipped with luxurious amenities, and surrounded by ancillary buildings were part of the Roman landscape in Bruttii during the later centuries of the Empire. If the theater at Marina di Gioiosa Jonica was truly part of the villa complex it would admittedly constitute a unique example in the region, and a rarity in Italy as a whole. Yet, private theaters were built in the villas of the emperors and in some luxurious villas on the Campania’s shore.423 As H. Mielsh points out, the diffusion of theaters in Roman villas developed when the emperors and members of the senatorial aristocracy acquired increased followings.424 It is likely that even in Bruttii the aristocratic families of senatorial rank had large entourages that visited and sojourned in their lavish residences, and that the ultra wealthy aristocrats of the region shared all kinds of luxurious amenities with their followers while ostentatiously displaying their wealth and refined culture. If we take into account the new archaeological discoveries, which point to a vital economy in the region between the third and fourth century AD, it is not to be excluded that the villa at Marina di Gioiosa Jonica contained a theater as well as private baths for the enjoyment of its owners and their visitors. Attributing the theater at Marina di Gioiosa Jonica to the villa or nearby center is only uncertain because of the unscientific excavations carried out when these findings were unearthed.425 Further exploration in the area of the villa using modern instruments and scientific methodologies may confirm Ferri’s hypothesis that the theater, the baths and the other scattered findings were indeed linked to the villa and its life. Whether the theater at Marina di Gioiosa Jonica was connected to a nearby statio or was a private structure that belonged to a luxurious late Roman villa, it certainly stands as a further proof of the vitality of the territory near Locri, and of the wealth achieved by the local community between the third and fourth century AD.

423 See the Villa on the island of Pianosa, where perhaps Agrippa Postumus went into exile; and the Villa Pausilypon, the luxurious home of Publius Vedius Pollio, one of ’s supporters. Upon his death the villa went to the emperor. This latter grandiose villa was composed of several buildings erected between the BC and the 4th AD, to cover an area of about 9 hectares. At the centre was the huge residential structure, around which were built a grandiose theatre, an odeion and thermal baths. Apart from its private dock on the Cala dei Lampi, the complex included its own roads and the tunnel known as the ‘Grotta di Seiano’ that crosses the rocky coastline to Coroglio, which today is the point of access to the complex. Other outstanding examples of theaters inside a villa are offered by the Villa of Domitian at Castelgandolfo and the Villa Hadriana at Tivoli. MIELSH 1987, pp. 108-110. 424 MIELSH 1987, pp. 108-110. 425 The mentioned findings were found scattered in the area and since their discovery they have been part of private collections. 148

2) The Villa of Naniglio in Gioiosa Jonica (fig. 6.4).

In Gioiosa Jonica (locality Annunziata), in the Torbido Valley,426 about 12 km. north of Locri and 100 km. north of Reggio, a remarkable monumental complex called the ‘Villa del Naniglio’ was discovered and excavated in the late 1980s. Its life spanned from the end of the first century until the fourth century AD, confirming continuity in occupation throughout several centuries.427 The villa was built 110 m. above sea level, in a splendid, panoramic position on the north-eastern slope of the valley crossed by the Torbido River. It is fairly close to both the Ionian Sea and the Torbido River, two natural features that must have played an important role in the villa’s development and the pursuit of its owners’ productive activities. The Torbido Valley had been occupied since prehistoric times. The valley communicated with the mountainous hinterland through which it was possible to reach the Tyrrhenian coast on the west side of Bruttii. Gioiosa Ionica in antiquity engaged in material and cultural exchange with the nearby important centers of and Locri,428 evidence that the villa must have benefited from the vicinity to such developed centers. The Villa of Naniglio was formed by a series of buildings distributed on three different natural terraces along a sloping hill.429 Scattered ruins suggest that the estate belonging to the villa extended up to the Torbido River. 430

6.4 Residential area on the lowest terrace of the Villa of Naniglio (3rd c. AD).431

426 The Greek and Roman name of this town is uncertain; see de FRANCISCIS 1988, pp.27-35. The Torbido Valley takes its name from the River Torbido that according to some scholars corresponds to the ancient river Sagra. This geographical area on the Ionic coast belongs to the modern province of Reggio Calabria and is situated between the Serre and Aspromonte ridges. It includes the modern towns of , , San Giovanni di , Martone, Gioiosa Ionica e . 427 For a comprehensive account of the villa’s excavations and findings, see: de FRANCISCIS, A., La villa romana di Gioiosa Ionica, 1988. 428 DE FRANCISCIS 1988, p.17. 429 The distribution of villas on sloping terraces along the sloping hill is found all over the Roman Empire and recurs in several instances also in Bruttii(see DE FRANCISCIS 1988, p.112 with notes). 430 See DE FRANCISCIS 1988, p. 21. 149

Between 1981 and 1986 the University of Naples excavated only the lowest terrace and unearthed an area that is 100 m. long. Crossed vaults and arches were uncovered that were probably used to create terraces on the sloping hill. In the southernmost area of the excavation (not visible in fig.6.4.), structures dated to mid-third century AD were uncovered. They were interpreted as part of a thermal facility separated from the residential area of the villa. The separation of the baths from the main nucleus of the building is a common characteristic in late Roman villas all over the Empire. In Bruttii, the villas at Palazzi di Casignana and at Quote San Francesco feature baths separated from the residential sectors.432 Besides the bath, the only building so far identified is located in the lowest terrace of the archaeological area (See plan fig.6.4). It is oriented north-east/south-west and features an elongated design with two projecting sections at each side. At its center an impressive water reservoir was found underground (fig.6.5). This facility was called “Naniglio,” a name used by scholars to refer to the entire villa complex.433

431 After DE FRANCISCIS 1988. 432 For a discussion on the baths of Villa at Palazzi di Casignana, see Chapter 4 of this study. The baths inside the Villa at Quote San Francesco are described in AVETTA-MARCELLI-SASSO D’ELIA 1991. 433 See ROHLFS, G. Dizionario toponomastico della Calabria, Ravenna 1974; FUDA, R., Naniglio: Etimo di un toponimo bizantino in Calabria, in Klearchos, 185, pp.33 ff. The name ‘Naniglio’ comes from the Greek anelios, ‘without sun’ which well describes the underground reservoir; for others the name derives from the Greek naos eliou, referring to a temple of Helios, see IACOPI, G., Il mistero del ‘Nanilio’ svelato? In La voce della Calabria, V, 174, 2 agosto 1947, p.1. This last interpretation was rejected by de Franciscis who argues against the hypothesis that the ruins belonged to a temple of Helios. 150

6.5. The underground reservoir.434

This water reservoir is of extraordinary dimensions (17,47 m. x 10,27 m.), and superbly engineered. It is one of the best preserved examples of an underground water reservoir discovered in Italy.435 The entrance to the hypogeum is located on the south side and features an arch that is 0,75 m. wide and 2,17 m. high. The reservoir possesses one central nave and two lateral aisles. The ceiling is cross-vaulted and supported by two internal rows of four quadrangular pilasters (each 0,90 m. wide). At the center of the reservoir there is a water-draining well. The walls are coated with water resistant tectorium while the floor is paved with opus signinum. On the western side of the reservoir there is a cylindrical tower. Originally it was roofed with a dome. Inside the tower there is a brick spiral staircase (see fig. 6.6) that rotates around a cylindrical pillar covered with bricks. The staircase starts from the roof of the reservoir and is composed of 24 steps which lead to the bottom of the room allowing periodical water inspections. This type of staircase is unique in Bruttii. It is attested in public buildings of Rome from the third century AD onwards (see the Thermae of Caracalla and the Basilica of Maxentius).436 The reservoir of the Naniglio has been compared to the Piscina Mirabilis at Misenum, located at the western end of the Gulf of Naples, because of its width and length (fig.6.5a).437

434 After de Franciscis 1988. 435 Today, the modern road S.S.216 that connects Gioiosa Ionica to Mammola runs on top of it. For other examples of water reservoirs, see DE FRANCISCIS 1988, p.24 and bibliographical notes. 436 For the Thermae of Caracalla, see PIRANOMONTE, M., The Baths of Caracalla, Milano Electa 1998; for the Basilica of Maxentius, see GIAVARINI,C. - AMICI, C. M., et alii, The Basilica of Maxentius: the monument, its materials, construction, and stability, Roma 2005. 437 BORRELLO, M.- D’, A., Baiae-Misenum, Regio I, XIV (Forma Italiae), Firenze 1979. See also, M. Döring, “Wasser für den 'Sinus Baianus': Römische Ingenieur- und Wasserbauten der Phlegraeischen Felder", Antike Welt 33 (3), 2002, pp. 305–319. 151

6.5a. Piscina Mirabilis at Misenum (Naples, Italy).438

In Bruttii, a remarkable villa located near Vibo Valentia, on the Tyrrhenian coast, in the small village of Zungri-Papaglionti, features a huge hypogeum that has been interpreted as a nymphaeum (fig.6.7). The villa surprisingly has not received much scholarly attention,439 but was admired in 1710 by a scholar who described the villa’s hypogeum as having a double staircase made of stone by which one could descend inside the double aisled hall.440 These water reservoirs are further proof that Bruttii utilized advanced hydraulic systems equal to that of Campania and central Italy during the late imperial period.

438 After http://en.wikipedia.org/wiki/Piscina_Mirabilis. 439 The villa is briefly described by ARSLAN, E.A., “La ricerca archeologica nel Bruzio,” in Brettii, Greci e Romani. Atti V Congresso Storico Calabrese (Cosenza-Vibo Valentia-Reggio C. 1973), Roma 1983, pp. 269-310; GUZZO, P.G., “Il territorio dei Bruttii,” in Societá Romana e produzione schiavistica, GIARDINA, A. and SCHIAVONE, A. (eds.), Roma-Bari 1981, p.134 with bibliographical notes; PAOLETTI, M., “Occupazione romana e storia delle cittá,” in Storia della Calabria antica: etá italica e romana, SETTIS, S. (ed.), 1994, pp.494-495 with notes. 440 BISOGNI, J., Hipponi, seu Vibonis Valentiae, vel Montisleonis, Ausoniae Civitatis accurata Historia, Neapoli 1710, Cosenza 1980, pp. 40-42. 152

6.6. Reconstruction of the hypogeum.441

441 After DE FRANCISCIS 1988. 153

6.7. Hypogeum of the Villa at Papaglionti (Vibo Valentia).

The water reservoir of the Naniglio is flanked by two small rooms that may have been used for service. The eastern room contains a terracotta aedicola coated with stucco. It is surmounted by an architrave that is supported by two semi pilasters (fig.6.6). The western room features a water pipe canal running along the southwest corner of the room. Both rooms have circular window for access of light and air. The long building and its two projections were likely the residential area of the villa. The southwest section features three rooms (fig. 6.4) that communicate through doors, all featuring floor geometric mosaics.442

442In the same area fragments of an opus spicatum have been found and remains of an opus sectile have emerged at east of the water reservoir. The existence of opus sectile floors is attested by a large number of marble tiles of various color and dimension found in a field east of the residential area of the villa. See, de FRANCISCIS 1988, p.53. 154

6.8. Drawing of Mosaic in Room A. 6.9. Mosaic of Room B.

The mosaics of Room A and B (fig 6.8 - 6.9)are in good conditions whereas the one in Room C is preserved only partially. All three are dated to the first half of the third century AD.443 The plan of a long building with two projecting sections at each opposite side does not recur in other areas of Bruttii, but it is quite frequent, even in its variants, across the Roman Empire. It is found in country villas especially outside Italy, in France, Spain, Africa, and Britain444 and corresponds to the typology of villas described in Varro and Vitruvius.445 Numerous fragments of colored plaster in a wide range of colors have been found in the debris of the excavated area, indicating the presence of painted walls. Marble thresholds, opus sectile fragments have been also found providing further evidence of the villa’s sumptuousness.446

Use of the reservoir Taking into account the huge dimensions of this hypogeum reservoir it is highly probable that it was build to provide water not only to meet the daily needs of the villa’s dwellers and the private baths, but also for the irrigation of the farming land connected to the villa estate and for brick manufacturing. The villa’s owners very likely were land owners involved in farming and agriculture in a very fertile area which even today still yields plenty of wine, olive oil and fruit.447The olive press discovered on the road that borders the villa complex, whose typology recurs in other villa contexts of Bruttii, would confirm that olive oil was produced at the Naniglio’s estate.448

443 A description of these mosaics and their motifs is in DE FRANCISCIS 1988, pp.53-63. 444 PERCIVAL 1976, p. 23, fig. 3 with related bibliography. 445 See Varro I, 13, 3; Vitruvius, VI, 6 ff. 446 DE FRANCISCIS 1988, p.41. 447 For the cultivation of grain and olive trees in Bruttii and Magna Graecia see, GAMBI, L., Calabria, Torino 1965, p. 298 ff; idem, 306 ff.; GHINATT, F., Aspetti dell’economia agraria della Magna Grecia agli inizi dell’Impero, in Critica Storica X, 1973, p. 369 ff. 448 de FRANCISCIS1988 p.17. 155

The production of olive oil has a long tradition in Bruttii, and is archaeologically attested in other fertile areas of the region.449 Another possible activity carried inside the Naniglio’s estate was the manufacture of bricks. To the east of the water reservoir archaeologists uncovered a small vaulted room; probably a water tank that, it has been suggested, was used for brick manufacture.450

The function of the villa

From the archaeological remains of the Villa del Naniglio, it can be stated with a fair degree of certainty that this building was the residential center of a vast praedium belonging to a well-off Roman family of the late imperial period.451The huge dimensions, elegant architectural style and technical expertise displayed in the construction of the water reservoir point to patrons of substantial wealth. They had to be economically capable of hiring the most skilled builders (perhaps from other regions of Italy), to install such a monumental reservoir, and experienced craftsmen for the realization of refined floor mosaics, and to sustain year long the high cost of water supply and the reservoir’s maintenance. One can only imagine the army of servants and work force necessary to run such an enterprise - that included not only the villa’s maintenance, the social gatherings of the villa’s owners, but also land farming on a vast estate which was likely combined with brick production. The Villa del Naniglio stands as a further remarkable testimony that an active upper class lived and prospered in Roman Bruttii between the third and fourth century AD. A complex so well planned that it solved the difficulties of the sloping terrain, was provided with so luxurious a facility as the private baths, and featured elegantly decorated residential space with floor mosaics, must have been owned by a very wealthy dominus with refined tastes. The villa of Naniglio even more attests that the upper class in this area of Bruttii was participating in the common trend of the Roman aristocracy in other regions of the Empire.

449 For instance is attested at Falerna, see LATTANZI, E., Klearchos, 1981, pp. 148 ff.; DRACHMANN, A.G., Ancient oil mills and presses, Köbenhavn, 1932. 450 de FRANCISCIS 1988, p.25. 451 On the extension, nature and production of praedia see SKYDSGAARD, J.E., Nuove ricerche sulla villa rustica romana fino all’epoca di Traiano, in Analecta Romana Inst. Danici, V, 1969, pp. 25 ff. 156

3) The Villa at Ardore

About 14 km. south of Locri, between Ardore Superiore and Ardore Marina, another late imperial villa was discovered in the locality of Salice.452 Sections of walls, squared blocks, sandstone column-drums and roof tiles were uncovered. Although much less is preserved here than the previous two examples, what does survives suggests that this was once a large and imposing villa. Guzzo dated this building between the third and fourth century AD, hypothesizing that it featured also a working farmhouse (pars fructuaria).453

6.10. Sarcophagus of Ardore (3rd-4th c. AD).454

A marble strigillated sarcophagus (fig.6.10), dated between the end of the third and the beginning of fourth century AD,455 was discovered in the necropolis near the villa.456 It is a high-quality piece of marble carving, the chest being ornamented with a portrait (the features are not carved but the hair style is that of a woman) and bucolic figures. It was evidently commissioned in Rome by the villa’s owner for a deceased woman of his family.457 The presence of an imported marble sarcophagus attests that the dominus of the villa was very wealthy, likely a result of his land-holding. The following section describes a villa that was discovered 5 km. south of Locri, at Quote San Francesco near the river , which shares similar characteristics with the Villa at San Giovanni di Ruoti in Lucania (modern Basilicata).458 Although this complex is dated to the fifth century AD, a century after Villa at Palazzi di Casignana’s renovation, it stands as significant evidence that well-appointed mansions were not lacking even later in time in this part of the Empire.

452 ACCARDO 2000, p.88. Part of the villa’s material was reused for the construction of the nearby Church of San Salvatore. 453 GUZZO, P.G. “Il territorio dei Bruttii dopo il II sec. d.C.”, in AAVV., Calabria bizantina, Roma 1986, pp.109- 120. 454 After FAEDO 1994. 455 C.Sabbione dates the sarcophagus to the fourth century AD, see COSTAMAGNA-SABBIONE 1990. L, Faedo, instead, dates it to the last quarter of the third century AD, see FAEDO, L., Aspetti della Cultura Figurativa in Età Romana, in Storia della Calabria, v. 2, 1994, p.627. 456 The presence of burials near late Roman villas and their significance is discussed in Chapter 9. 457The sarcophagus’ decoration is treated in FAEDO L., Aspetti della Cultura Figurativa in Età Romana, in Storia della Calabria, v. 2, p.627; for a report on the residential villa at Ardore see ACCARDO 2000, p.88, n.14. 458 SMALL- BUCK 1994. 157

4) The Villa at Quote San Francesco

6.11. Villa at Quote San Francesco: plan (5th c. AD).459

The Villa at Quote San Francesco is dated to the fifth century AD and marks a change in the development of Roman domestic architecture. This building is placed within the tradition of the Roman villa, yet it departs from that tradition in its compound arrangement of space: the upper story used for residential purpose, and the fortress-like character of the architecture. All of these elements essentially mark an important transitional moment. The villa is composed of two separate sectors, revealing a multifunctional use: the baths and a residential area (fig. 6.11). The current state of excavation does not allow a decision as to whether these two areas of the complex were somehow connected through a corridor or portico when the complex was built. The baths feature small rooms distributed around a central circular room (A) (a tepidarium?) that features suspensurae under the pavement; the walls of the room have four niches that were originally painted. The archaeologists do not suggest what their use was, but perhaps they were intended for statues. Fragments of painted plaster and marble tiles found in the room’s debris suggest that this space was once elegantly decorated. One tub (E) is located to the south-west of Room A, and a second one (D) to the south-east. Rooms F and B were also heated, as the presence of suspensurae confirms. Room C featured tubuli within the walls, thus it too was heated. Room F contained scattered marble tiles that decorated the floor. Another tub (I), and a small room (H) completed the thermal sector of the complex. The residential area has not been completely excavated but extended for 60 m. in length. The eastern side of the residential area features a room with a polygonal apse (J), which is 20 m. long and had three arches; on the south side of Room J lie two rooms (K and L), whose walls show aligned holes in which were inserted the beams that supported an upper story. The western side of the building, the one near the baths, features an apsidal space symmetrically opposite to Room J. The space in between the two opposed apsidal rooms was perhaps occupied by a courtyard but more excavations would be needed to confirm this hypothesis. On the south side of the building there is a long rectangular space, perhaps a portico. The residential area of the complex was organized in two floors as suggested by holes for supporting beams along several

459 After ACCARDO 2000. 158

surviving portions of walls. It is assumed that the upper floor was used as a living space whereas the ground floor, having a low ceiling, no windows, and no decoration at all, was likely used as either a stable for horses or as a service area of the complex. The complex at Quote San Francesco has been compared to the villa at San Giovanni di Ruoti in Basilicata.460 Both buildings feature similar characteristics: 1) A compact architecture created with different sections put side by side rather than arranged around a peristyle; 2) They contain baths and apsidal reception rooms; 3)The residential areas are located on the upper level. This element represents a novelty in villa architecture and it was found in other contexts: at the villa of Galeata near Ravenna, which has been attributed to Theodoricus or to a member of his court; at the Palazzetto di Monte Barro in Lombardia, which is a small palace composed of three sectors arranged along three sides of a peristyle.461 It is possible that the transfer of the residential area to the upper story at both Quote San Francesco and San Giovanni di Ruoti was determined by defensive reasons during times in which bandits constituted a serious threat to social stability.462 Although it would have to be confirmed by further excavations, the Villa at Quote San Francesco may also have featured a tower, as the Villa in San Giovanni di Ruoti does, making the building a defensible fortress. The building at San Giovanni di Ruoti has been defined as a “praetorium”, meaning a power-seat from which the surrounding territories (vici) were controlled. 463 According to Sfameni, the same definition can be applied to the complex at Quote San Francesco.464 The term ‘praetorium’ in late antiquity replaces the more generic term ‘villa’; it refers to a residential building or the residential area of a large complex.465 Palladius (Pall., I, 8) names the building of the dominus his ‘praetorium’ and gives instruction on where and how it ought to be, underlining the importance of the building’s location in a elevated area in order to enjoy a pleasant view rather than defend and control the surroundings. The pars rustica and the pars fructuaria are separated from the praetorium, the residential area. Praetorium seems here to replace the expression ‘pars urbana’, or ‘villa urbana’ used by earlier Latin authors of agriculture. Palladius in his Opus Agriculturae (I, 8), uses the word praetorium to refer to villas. Cassiodorus writes that the aristocratic possessores of Italy spent period of leisure (otium) in their suburban residences that he calls praetoria. (Variae VII,5,4; XI,14,3; XII,22).466 In short, the building at Quote San Francesco may be called either a praetorium or a villa – both terms convey the same idea of luxurious suburban residence.

460 See SFAMENI 2004, pp. 609 ff.; for a comprehensive account of the villa at San Giovanni di Ruoti see SMALL, BUCK 1994. 461 The relationship between late roman villas and medieval houses in general remains open due to scant material records. On the end of the Roman peristyle house see ELLIS 1988. 462 SMALL, BUCK, 1994, p.4. 463 SMALL 1998, p. 342. 464 SFAMENI, C., “Le villae-praetoria: i casi di San Giovanni di Ruoti e di Quote S. Francesco,” in Paesaggi e insediamenti rurali in Italia Meridionale fra Tardoantico e Altomedioevo, Atti del Primo Seminario sul Tardoantico e Altomedioevo in Italia Meridionale (Foggia 12-14 febbraio 2004), pp. 609-622. 465 Besides the lexica see UGGERI 1983, 318-320; in particular see note 20 for the use of the term meaning ‘stationes’; see RODA 1981, p. 131 for sources and bibliographical references. Modest. Dig.31, 1, 34; Palladius I, 8; Cassiodorus. Var. XI, 14; XII, 22. For the use of the term in regards to the Villa at Desenzano, see SCAGLIARINI CORLAITA 1990, 257; Modest. Dig.31, 1, 34; Palladius I, 8; Cassiodorus. Var. XI, 14; XII, 22. 466 Praetorium is also the seat of a military command; at times it designates the stationes, the Roman posts in which travelers would refresh and feed their horses. Praetorium also means the residence of the emperor. See SFAMENI 2005. 159

To Summarize Since the Republican period, in Bruttii Roman villas were linked to large-scale agricultural production. The villas near Locri so far examined prove a continuation of this agricultural tradition even into Late Antiquity. The four villas so far described do not rival the grandiosity of the Villa at Palazzi di Casignana neither the opulence of its decorations. Yet, they all clearly belong to the same category of luxurious country and suburban residences that future excavations may uncover more numerous in Bruttii. There are still too few late Roman villas in Bruttii to allow scholars to define a villa typology for this region.467 Yet, the large and luxurious villas of late Roman times discovered in the area of Locri reveal remarkable agricultural and industrial productivity, and a considerable concentration of wealth in the region.468 The presence of private baths, which entailed a high and expensive consumption of water, together with the grandiose layout, technical complexity and decoration found, for instance, at the villa of Naniglio, are some of the significant expressions of the economic strength and dynamism of the Roman elite of Locri and surrounding areas between the third and fifth century AD. If, as I expect, the attribution of the Roman theater to the villa of Marina di Gioiosa is confirmed by deeper investigation of the site, we will have an ostentatious manifestation of the wealth reached by some late imperial aristocratic families in Bruttii during a time that was once considered economically depressed and archaeologically unimportant.469 This, in turn, will invalidate preconceptions and assumptions about Roman Calabria that are still widespread in the field.There is no evidence to help us to attribute any of these residences to a particular aristocratic family who lived in the territory of Locri between the third and the fifth century AD. Yet, several inscriptions dated to the early imperial period record that important families had their possessiones in the area of Locri.470 In the third century AD the Aurelii and Caerellii for instance had possessions in Gerace, the Cornelii and Uagellii near Locri, and the Duc(eni)in Ardore. 471The presence of these families and their villas in the region forces us to reconsider the negative cliché about Bruttii expressed, for example, by Seneca the Younger (4 BC- 65 AD): Inculta uideantur, Bruttios et Lucaniae saltus persequamur.472 The villas of Bruttii were both productive and residential centers tending to decrease in number over time while becoming more monumental in comparison to their earlier counterparts. These villas here examined attest to the prosperity and enterprise of the local, municipal elite whose members surpassed one another in ostentation precisely in the same manner as it appears in other regions of the Empire.473

467 The problem was already addressed in ARSLAN 1983, pp.271-310; NOYÈ 1988, p.73. 468 Other notable Roman buildings in Locri are a complex in Contrada Petrara, perhaps a public basilica dated to the second century AD, and the baths of a late imperial Roman villa incorporated in the 1800 residence of Casino Macrì. This complex had its own aqueduct, the arches of which are still extant. See COSTABILE 1976, pp.117- 118. 469 This view was challenged for the first time by KAHRSTEDT 1960, pp.121 ff. 470 COSTABILE 1976, pp.81-85. For a treatment of the inscriptions that attest to the presence of local aristocratic families see SEGENNI, S., “Economia e società in età romana: la documentazione epigrafica,” in SETTIS 1994 (ed.), pp.655-667. See also COSTABILE, F., “Dalle poleis ai municipia nel Bruzio romano,” in SETTIS 1994, pp.439-464, especially pp.455-457.. 471 FIACCADORI 1994, Storia della Calabria, p.738. 472 Seneca, De Tranquillitate Animi, 2.13. 473 According to F. Costabile, the gentes of Locri, Rhegium, and other major towns of Bruttii, attested by inscriptions, were originally from Central Italy. Under the reign of Augustus, several Roman families immigrated in Bruttii and contributed to the Romanization of the region. Some of these gentes as, for instance, the Uagelli, settled 160

CHAPTER 7: Rural Estate and Urban Center in Late Antiquity.

When the Villa at Palazzi di Casignana expanded to its greatest size in the fourth century AD, other villas reached a similar level of monumentality in several different areas of the Western Empire. The fourth century was a period of renaissance for elite villas marked by two parallel phenomena: 1) a significant increase of size and opulence; 2) a substantial reduction in their number. New villas were designed at a larger scale. If a villa had been built in earlier times and it was partially or completely renovated, then its residential structures were endowed with monumental architectural form and lavish decoration. 474 On the whole, these fourth century villas employed apsidal and curvilinear spaces, large reception rooms, vestibules and courtyards at the entrance, fountains, gardens and thermal facilities. As Y. Thébert points out, an international character developed among the upper classes of the late Empire in regards to their domestic architecture.475 A wave of studies that departs from the thesis elaborated by the Russian historian M. Rostovtzeff links the drastic drop of the number of villas to a general regression of the agriculture and manufacture production of Italy and other regions of the Roman Empire during the ‘third century AD crisis.’ According to A. Carandini, the third century crisis was due to several factors, including the decrease of the slave market for lack of new conquests, the competition generated by farming products of the Roman provinces such as Africa and Spain, and the incapability of reforming the agricultural system, which was trapped in rigid structures and based prevalently on the exploitation of latifundi whose exaggerated growth became in the long run a cause of weakness.476 For all of these reasons, between the second and the third centuries AD, many villa owners fell into poverty and were forced either to abandon their residences with the annexed estates or to sell them to more successful land owners. The possessores who sustained their wealth through the crisis, on the other hand, took advantage of the critical economic situation. They bought out or incorporated the abandoned properties left by impoverished landlords. By doing so, they were able to invest in land and multiply their productivity. From the new wealth that they accumulated by successfully managing their estates, they were able to build their enormous villas or renovated them with greater opulence and monumental style than in earlier times.477 In Bruttii, between the second and third century AD more than 60% of the villas that had flourished in the Republican and early imperial period,- and that were once propulsive centres of rural economy disappeared. This phenomenon continued during the fourth and fifth centuries affecting principally the coastal areas of the region. In fact, only 8% of the Roman villas so far discovered in the region belong to the late imperial period.478 They are typically located on the

in several towns including Locri, Vibo and Rhegium since the first century until the fourth century AD. Based on the inscriptions, these families were generally involved in the local public life. COSTABILE, F., Municipium Locrensium, Reggio 1976, pp. 75-85. 474 A large survey on rural residential architecture is in SODINI 1995 and 1997. A general account on the urban and rural architecture is in HIRSCHFELD 1999. For changes in pre-existent villas see PERCIVAL 1989, p.284. On the late antique countryside see CHAVARRIA-LEWIT 2004; CHRISTIE 2004. 475 THEBERT, Y., “Vita privata e architettura domestica nell’ Africa romana,” in ARIÈS, P., DUBY, G., La vita privata dall’Impero romano all’anno mille, Roma-Bari 1986, pp.233-309. 476 CARANDINI 1995. 477 In regards to Bruttii this view is held by SANGINETO 1994, pp. 584-585. For Apulia, see VOLPE 1996, p.210. 478 SANGINETO 1994, pp.584 ff. 161

coast where they could take full advantage of both the sea and the fertile valleys and their impressive landscape. These villas are also situated a short distance from the major towns, and each one is located very near the principal or secondary Roman road networks. In contrast to the predominant view of the third century crisis (which supposedly involved also Bruttii) new discoveries of lavish villas in several regions of the Empire, which are dated to this period, are motivating scholars to talk more about ‘transformation’ rather than crisis, lowering the severity of the economic changes occurred in the third century AD.479 An important prosopographical study by F. Jacques, demonstrates that the senatorial order was not affected by the ‘third century crisis.’480 Their country villas at this time were, in fact, rich, monumental and furnished with all kinds of comfortable amenities typical of the urban way of living, such as the baths (which sometimes are even doubled). They were places for recreational activities, cultural pursuits, physical exercise, bathing, and entertaining guests. But they were also the hub of a working farm, -although this might not always be the case since some maritime and suburban villas were built more for pleasure than profit. The ‘third century crisis’ has been traditionally assumed to be connected to, or have caused, a progressive decline of towns.481 Many aristocrats would have had close financial interests in the country, which supposedly motivated them to abandon the towns, and to migrate into the country in order to live there permanently.482 These late Roman landlords would have spent most of their time in their villas, managing properties, collecting tributes from tenants, and exercising patronage and legality over their homines.483 In brief, the late imperial landowner would have been a sort of proto-feudal lord who lived on his estates supervising his regna. But even this picture is not completely accurate. It is a fact that several excavated villas feature both large, productive facilities and comfortable residential areas, proving that the economy of these complexes was in large part self-sufficient. And it is also true that the presence of artisans, bishops and all kinds of work force on late Roman estates attested by literary texts (at times corroborated by material findings) point to the autonomy of late Roman villas.484 But the notion of impoverished and abandoned towns versus flourishing countryside populated by wealthy villas should be mitigated by recent archaeological discoveries and studies on urban areas in several regions of the Empire.485 The new ruling classes formed by the imperial burocracy that emerged after the Diocletianic reform, while invested in their rural properties, continued to participate in the urban life as proven by the discovery of well appointed town

479 See, for instance, VERA 1994b; LEWIT 1991; LIEBESCHUETZ 2001; DE BLOIS 2002. 480JACQUES, F., “L’ordine senatorio attraverso la crisi del III secolo,” in Società romana e impero tardo antico, vol.1: Istituzioni, ceti, economie, in A. Giardina (ed.), Roma-Bari 1986, pp.81-125. See also CRACCO RUGGINI 1998; GARBARINO 1988; SALZMAN 2002, pp.29 ff. According to Salzman, the third century senatorial aristocracy retained its role as social and economic elite in the state. 481 BURNS, T. S.- EADIE, J.W. (ed.), Urban Centers and Rural Contexts in Late Antiquity, 2001. 482 KEAY 1988, p. 190. 483 For an account of the scholarship that portrays the late Roman dominus almost as a feudal master see the bibliographical references in CRACCO RUGGINI 1961, p.85; For Italy see DOREN, A., Storia economica dell’Italia nel Medio Evo, Padova 1937, p.28; CARLI F., Il mercato nell’Alto Medioevo, Padova 1934, I, pp.43 ff.; ROSTOVZEV 1933, pp.530-531. For Gallia see PIGANIOL, A., L’empire chrétien, Paris 1947, pp.361 ff.; for Britain see COLLINGWOOD, R.G., Roman Britain and the English Settlements, Oxford1936, pp.209-220. 484 See, Vita Melaniae and Palladius, Opus Agriculturae. 485 CAGIANO DE AZEVEDO 1986; RICH 1992;BROGIOLO-WARD-PERKINS 1999; BROGIOLO- GAUTHIER-CHRISTIE 2000. 162

dwellings.486 Until recently, the late antique city was seen largely as a weak continuation of the classical Roman town. Yet, in the last thirty years, new fieldwork has led to a revision of negative stereotypes about the period. New excavations have shown that between the fourth and sixth centuries AD cities in the Central and Eastern Mediterranean flourished. Yet, little research has been done in regards to how buildings and public spaces were used in daily life, although the general picture of urban development in this period is better known than in the past. Topographical studies have tended to concentrate on the later history of private architecture and churches, leaving out public spaces including streets, fora and secular buildings. In the last few years archaeological surveys in several parts of the Empire including Spain487 or Southern Italy,488 have brought scholars to reconsider the view of abandoned urban centres and the loss of their leading economic role.489 Findings indicate that many towns across the Empire continued to represent important administrative and political centres even in the late Roman period.490 Textual data support the archaeological evidence and reveal that the city continued to be the preferred place of residence for the notables of the provinces and that rural potentiores were also the urban elite.491 In several instances archaeological data reveal both a reduced population and less building activity in urban areas compared with the early imperial phases. Yet, this phenomenon does not seem consistent and requires much more attention towards each territory in every Roman province in order to develop data beyond schematic generalizations.492 In short, the movement towards the country is not necessarily to be taken as a sign of urban crisis. It could have been dictated simply by the convenience of controlling and administrating the agricultural economy practiced by the elite. The economic phenomenon outburst of the villas in the fourth century AD had also a political element that may indicate the presence, at least in the major Roman urban centres of Italy, of an administrative and political body. Villa owners in many instances likely had a role in the administration and public affairs of the town. Landowning families and those who did not supervise work on their estates were encouraged to live in their municipia and to take part in the administration.493 A law by Valentinian in 364 AD prohibited the provincial governors to exercise justice in their private residences (secessu domus), providing important evidence that urban centres were still considered the main political and administrative seats in the fourth century AD even if the tendency of some officials was to exercise part of their function in their homes.494 Taking all of this into account, it is likely that the opulent villas built in the vicinity of important towns such as Villa at Palazzi di Casignana near Locri, Villa at Desenzano near Verona, and Villa at Faragola near Herdonia, may have been the residences of the city officials and would thus confirm the vitality of the urban centres in the fourth century AD rather than

486 GUIDOBALDI 1999 (Rome); BALDINI-LIPPOLIS 2001 (Mediterranean); CHAVARRÍA 2005a; RIPOLL- ARCE 2000 (Spain). 487 For the Late Antique villas of Spain see CHAVARRIA-ARCE-BROGIOLO 2006. 488 VOLPE 2004. 489 LAVAN 2001a; DE BLOIS 2002. 490 DIAZ 2000; KULIK 2000. For an analysis of the cities between Late Antiquity and the Early Middle Ages see WARD-PERKINS 1996. 491 GARNSEY, P.- WHITTAKER, C.R., “Rural life in the later Roman empire” in The Cambridge Ancient History Vol. 13: The Late Empire, A.D. 337–425,; idem, “Trade, industry and the urban economy,” 1998. 1990, p. 111; see also VERA 1983, p. 491. 492 For a strong argument against the role of the city during the Late Empire, see LIEBESCHUETZ 2001. 493 See GEROV, B., Land Ownership in Roman Thracia and Moesia, Amsterdam 1988, p. 94. 494 Codex Theodosianus, De officio rectoris provinciae, Leg.2, Lib.1. –T. 163

their abandonment. M. I. Rostovzev in 1933 maintained that during the late Empire cities and countries constituted two contrasting realities where the country ended up prevailing over the city.495 By the 1960s, his view was disputed by S. Mazzarino, who argued that Rostovzev’s theory ought to apply only to small towns and not to main urban centres across the Empire. These remained vital and active even during the late Empire, constituting the natural frame of the rich country domini.496 Even literary sources attest to the vitality of urban centres at this time. They very rarely mention important rural areas. If they do, it is usually to denigrate the lifestyle of those nobles that chose to live there.497 Both Symmachus and Sidonius, for instance, in their letters invite their noble friends to return to the city and not to indulge in leisure time at their country estates.498 According to the sources, cities such as continued to be highly populated and bursting with traffic and activities.499 Turin was splendid and full of entertainment.500Verona, and all the main cities of Venetia, were all flourishing with industry, commerce and public architecture.501 In contemporary texts rich late Roman possessores are often portrayed as managing their lands from towns leaving to villici and actores the management of their rural properties. These aristocrats go to religious ceremonies on the letiga or horse, dressed up with silk garments, adorned with jewels and followed by a long line of servants.502 Rural settlements all over the Empire are still little known due to the difficulty at identifying them and to the modest and perishable material of their buildings.503 Deep territorial investigations across Bruttii are needed in order to have a better picture of the relationships between country and towns in late antiquity. A better knowledge on the subject would also benefit an understanding of the phenomenon of monumentality that concerned many late Roman villas including the one at Palazzi di Casignana. In Bruttii, as fig. 7.1 illustrates, late Roman villas as well as their earlier counterparts, clustered near major towns, strongly indicating the importance of towns, even at the end of the imperial time. Starting from the north of Bruttii, there are notable concentrations of villas around Thuri/Copia, Croton, Scolacium, Locri, Rhegium and Vibo Valentia. Typically villas are located at an approximate 15 km distance or the equivalent of a day’s journey from a town.504 The villas in the territory of Locri described in Chapter 6 are also at approximately 15 Km. away from the town and about 100 km away from Rhegium, emphasizing the importance of

495 ROSTOVZEV 1933. 496 MAZZARINO 1951, pp.252 ff. On this subject see also CRACCO RUGGINI 1961, pp. 81 ff. and more recently CRACCO RUGGINI 2004. 497 This is a very long-standing literary tradition. See ROSEN-SLUITER 2006. 498 Sidonius, Epistulae, 8.8.2. An analysis of Symmachus’s letters that show the late Roman senators’ attitude towards otium in their suburban villas, see RODA 1985, pp. 101-106. 499 Ausonius, Ordo nob. Urbium V, Mediolanum: the city is said to be second in importance after Rome. See CRACCO RUGGINI 1961, p.83. 500 MAXIMUS TAUR., Opera, P.L. 57, passim; CRACCO RUGGINI 1961, p. 83. 501 CRACCO RUGGINI 1961, pp.83-84 and related bibliography. More recently, CRACCO RUGGINI 2004. 502 Ibidem, pp.85-86 with references to the ancient sources. 503 BROGIOLO-CHAVARRIA 2007, pp.23-30. 504 Different scholars use different measurements for this. See, R. Hingley, Rural Settlement in Roman Britain (1989), pp.112-114 argues for a 10-15 km range for standard half-day’s journeys (i.e., any two sites within 10-15 km can be visited and returned from on the same day – i.e., a total journey of 20-10 km); T. Bekker Nielsen, The Geography of Power: Studies in the Urbanization of north-west Europe (1989), p. 32, argues for 20 km. 164

transport and access to and from these rural sites. In Bruttii, the major towns of the region show a high degree of vitality in Late Antiquity.505

505 G. Noyè argues that Bruttii during late antiquity saw the rise of economic activity in rural areas while the vibrancy of many urban centres waned. NOYÈ, G., Economia e società nella Calabria bizantina, (IV-XI secolo, in A. Placanica (ed.), Storia della Calabria Medievale, Roma-Reggio, pp.577-655. 165

166

7. 1. Distribution of Roman villas in Bruttii.506

Starting from the north of the region, Thurii-Copia (the heir of the Greek town of Sibaris) is still an important port on the Ionian Sea and will be abandoned in sixth century.507 Scolacium, the native town of Cassiodorus, had a forum of exceptional size that was still used through the age of Honorius (408-422 AD). The town has provided evidence of occupation until the seventh century AD.508 Vibo Valentia, on the Tyrrhenian coast, remains populated and lively until the end of the Roman occupation;509 Rhegium located at the tip of the Italian peninsula, bridging the Ionian and Tyrrhenian Sea, was an obligatory military and civil port that connected traffic among Rome, Sicily and the East. Seat of the corrector of Lucania et Bruttii from the third to the sixth century AD, Rhegium became seat of the rector patrimonium of Bruttii during the time of Gregory the Great (540 c.a. – 604).510 Its large population stimulated an important market for olive oil and wine as documented by both local and imported African amphorae found in large quantity in the city. 511 Although archaeological data are still insufficient to reconstruct an exact organization of the Roman town in general, also Locri must have been a vital centre in the fourth century AD if between the third and fourth century AD the Roman theatre, located just outside the town, endured architectonic transformations. The cavea was reduced in size, the arena enlarged, and three small rectangular rooms were added.512 This proves the existence of a dynamic population and an effective city’s government at that time that invalidates the traditional view of the town falling in decline and losing importance at the end of the Roman age.513 Moreover, Locri became a in 330 AD. Christian inscriptions dated to the third and fourth century were found in Locri and in the nearby towns.514 The recent discovery (July 23, 2003) in Contrada Petrara at Locri of a fragment marble statue representing a magistrate further attests that the town was vital.515 Locri was inhabited until the fifth and sixth centurires AD after which the population moved to Gerace located on the hills for better defence from the barbarian attacks. The suburban villas discovered in the area of Locri show the degree of wealth and prosperity reached by the late Roman upper class that most likely lived in the town.516 Locri had a sea port that perhaps was connected with the river Torbido.517 This river today is just a creek running through the ‘Valle del Torbido,’ but in antiquity it was likely navigable constituting a sort of isthmian way

506After SANGINETO 1994, p. 576, fig.17. 507 FIACCADORI 1994, p.711. 508 Ibidem, pp.710-711; ARSLAN 1991. 509 Part of the thermal facility inside the complex named ‘Sant’ Aloe’ was transformed into a Christian church in fourth century AD. South of Vibo, at the Castle of Bivona, a mosaic dated between the fifth and sixth century AD featuring a Christian inscription was uncovered. FIACCADORI 1994, p.733. 510 Ibidem, pp.727-728. 511 Variae, XII, 14; NOYÈ 1998, p.448 512 see SABBIONE, C., Una città in Magna Grecia. Locri Epizefiri: guida archeologica, 1990. The theatre was originally built by the Greeks and throughout the Roman times likely hosted ludi and venations. 513 U. Kahrstedt’s pessimistic view in regards to the decline of Locri during the late imperial period is argued by ARSLAN 1983, pp. 269-310; PAOLETTI 1994, p.511; and more recently in D’AGOSTINO 2004. See KARDSTEDT 1960. Inscriptions provide the most information on the Roman municipium, see COSTABILE, F., Municipium Locrensium 1976. 514 COSTABILE 1976. 515 D’AGOSTINO 2004, p.41, note 4. 516 See Chapter 7 of this study for a discussion on the municipal elite. 517FIACCADORI 1994, pp.739-740. 167

between the Tyrrhenian and the Ionian coasts, facilitating trade in and out of Locri.518 This is confirmed by a large number of late Roman ceramic findings in the Valle del Torbido, including the ancient town of modern Gioiosa located about 10 km from Locri. Many imported African sigillata and amphorae were found scattered all over Gioiosa’s territory. Moreover, the presence of wealthy suburban villas as the Villa of Naniglio and the one in Marina di Gioiosa Jonica described in chapter x. attest to the importance of this urban centre and its involvement in commercial trade during the late imperial time. At the Villa at Palazzi di Casignana, luxurious material such as marble veneers imported from Greece and Africa that was used to decorate the entire residence point to continuity in the exchange relationship between the countryside and the nearby towns of Locri and Rhegium at least until the middle of the fourth century AD.519 In order to furnish their villas with extraordinary material, the villa’s owners must have relied upon the services and a network of people offered by these two influential centres on the Ionian coast. Based upon these observations, the picture of Villa at Palazzi di Casignana in the fourth century AD as an autonomous centre inhabited by its aristocratic owners who abandoned the waned nearby town of Locri does not seem suitable. Locri was likely a vital centre at the time when the villa developed. Being only 15 km distant from Villa at Palazzi di Casignana, it is possible that Locri played an important role in the economy and life-style of the villa owners. This, in turn, encourages us to further explore the late Roman phases of this important ancient town.

518 The frequent topographical names of ‘Crini’ in the entire area of the Valle del Torbido likely derived from the Greek term xrene meaning ‘spring’ and seems to refer to abundance of water. See ibidem, p.740. 519 In other Roman areas this pattern is documented in ancient sources. See… 168

Villa at Palazzi di Casignana and the mobility of late Roman elite

Another question to explore is whether Villa at Palazzi di Casignana, and late Roman villas in general, were permanent residences. The luxurious character of late Roman villas does not necessarily indicate that the late Roman elite lived permanently in these residences after having abandoned their urban houses and retired from the public life and affairs.520 In fact, it is possible that Villa at Palazzi di Casignana, as other contemporary villas of the Empire, was one of the many estates owned by a wealthy fourth century owner who may have visited this estate occasionally. It was customary for late antique possessores to delegate intermediary figures, conductores or coloni, to administer properties, as was the case with the estates of Symmachus in Sicily.521 In one letter Symmachus laments that the administrators of the properties do not fear their landlords, who are absent and far away. 522 Symmachus owned estates in Sicilia and Southern Italy but does not mention in his letters any stopover in these properties. He recalls only one trip to visit his properties in Africa.523 He administrated directly his properties in the Campania-Latium areas because these were close to Rome, where he held important magistracies. Whenever he resided in his country villas, he combined leisure (otium) with (business) negotium.524 Paulinus of Pella writes that he was never able to visit in Achaia and some estates that he inherited from his mother.525 In Chapter 8 of this study we will see how widespread the aristocratic patrimony of rich families was in late imperial times. Since the early imperial time, the custom for the Roman elite was to possess several estates scattered in different regions.526 This phenomenon reached a in fourth-fifth century AD. Literary texts recount that many aristocrats owned properties scattered across the Empire (in Iberia, Aquitania, Africa, Sicilia etc.). They resided in their country villas for short periods of time rather than permanently, visited their villas occasionally, or ever at all. These villas, while being self-sufficient establishments that often generated profit, were also expressions of the upper classes’ political asset. Late Roman villas played a role in the expansion of social networks among prominent aristocrats across the provinces of the Empire. Senatorial aristocrats who wanted to stay competitive had to cultivate relationships with influential people beyond the local politics of their native cities. They pursue prestigious political careers through the patronage of powerful officials with influence at the emperor’s court. In order to establish or reinforce important political connections, members of the Roman elite had to travel between cities and at times even across whole provinces. Travel and epistolary communication were almost obligatory practices in fourth century for prominent people who wished to sustain their wealth, status and power.527 Thus, it became essential for the elite to own many villas in different locations so that they could be ‘at home’ in dignified fashion even when they travelled. It was also important to provide their respected political guests and friends with luxurious amenities and all kinds of

520 ARCE 1997 b, pp.23-24. 521 Symm. Epist., 6, 11; 7, 66; 9, 52. On the agrarian system of Late Antiquity see VERA 1994c; WICKHAM 1994, 2002. On the coloni of fourth century Africa see VERA 1987. 522 Symm. Epist., 9, 6. 523 VERA 1986, p.260. 524 VERA 1986, p. 258-259; Symm., Epist., 1, 5, 2-10; 2, 60; 3, 23; 6, 66; 7, 18. 525 PAULINUS OF PELLA, Eucar., 413-419. 526 See Seneca, Ep. 87,7: dives …in omnibus provinciis arat. 527 See ELLIS, L.- KIDNER, F., Travel, Communication and Geography in Late Antiquity, 2004. 169

comfort. The villa became a power center and meeting point of important social networks and political relations during late imperial times. Therefore, the monumentality and lavishness of many villas reflects not only the wealth of their owners but also the important social role that these residences played in the creation of influential networks. Prominent families needed to forge and extend connections of friends and allies. They did so by holding important offices and by cultivating friendship. The aristocratic world that is constantly in motion and constantly networking is well documented in both Libanius’ and Symmachus’ epistles.528 The late antique potentiores remodelled and enhanced their villas to deliver a message of power. In emulation of the emperors and taking inspiration from the architecture of their residences, such as the Constantine’s Aula Palatina at Trier;529 the Maxentius palace-villa complex on the outside Rome,530 and the imperial villa of Pardorf, the late Roman elite built their villas as “stages for elite competition.”531 Even at Palazzi di Casignana, the villa’s expansion during fourth century AD is contemporaneous with an outward display of opulence. The villa developed with the addition of a monumental residential area, which included the very elaborate apsidal Room I, costly pavements, mosaic floors and lavish marble veneers. Presumably lost wall paintings, hanging pictures and statues furthermore enhanced the villa. The display of wealth might have been so ostentatious that it conveyed a sense of supremacy to the viewer. It is very likely that even the owner of Villa at Palazzi di Casignana was influential and a prominent political figure either within the political and social facies of Locri or more generally in the province of Lucania et Bruttii, if not from an other province. The villa may have been not only a retreat residence, but more importantly a place for conviviality and aristocratic gatherings, as the presence of baths and reception rooms indicate. By enhancing this residence and boasting his huge wealth, the dominus of Villa at Palazzi di Casignana asserted his political influence over the territory and its community.

528 See BRADBURY, S., “Libanius’ Letters as Evidence for Travel and Epistolary Networks among Greek Elites in Fourth Century,” in ELLIS - KIDNER (eds.) 2004, pp. 73-80; in the same book see SALZMAN, M.R. “Travel and Communication,” pp. 81-95. 529 ZAHN 1991. 530 IOPPOLO-PISANI SARTORIO 1999. 531SCOTT 2004, passim, pp. 39-65.

170

CHAPTER 8: The Economy of the Fourth Century Roman Elite and that of Bruttii.

Introduction

What were the social and economic conditions in Bruttii when the Villa at Palazzi di Casignana was expanded and embellished to monumental scale during the fourth century AD? What type of economy produced the kind of enormous wealth that is exemplified by this villa? How do the living arrangements of the late Roman elite residing in Bruttii compare with those living in other parts of the Empire? The next two sections explore the literary and material evidence that help answer these essential questions. The first section provides a general overview of how the Roman elite, including that of Bruttii, achieved enormous wealth during the Late Empire. It examines two parallel aspects that economically empowered the Roman upper classes: the accumulation of land widely distributed across the Empire; and the huge fortune that was also amassed from trade and commerce. A brief description of the monetary economy, based on the golden solidus introduced by the emperor Constantine, further shows how this reform significantly increased the economic power of the late imperial elite and widened the gap between the upper and lower classes of the Roman society. The second section looks more closely at Bruttii, its economic growth in the fourth century AD and the possible reasons that prompted this development. It outlines the available natural resources that very likely stimulated the economy of Bruttii, by encouraging the Roman nobility to invest their capital and efforts in this region during Late Antiquity, even more than they had done in the late Republic and the early imperial period. Ancient texts offer important clues about the abundance of natural supplies found in Bruttii and how the Romans exploited these resources. The authors of these texts wrote during different periods of Roman history: (AD 23-79), Cassiodorus, (c.AD 490-c. 585 AD), and Gregory the Great (AD 540-604). These writers describe various natural and geographic factors of the region, which likely contributed to a successful economy during the fourth century AD. The presence of a luxurious villa such as that at Palazzi di Casignana does not sufficiently prove that wealth spread across the whole Bruttii during the fourth century AD, but the discovery of other late imperial villas near Locri provides evidence that the elite invested significantly in developing this agriculture-rich region. Some information about Bruttii given by ancient authors has been confirmed by archaeological data, but further exploration of modern Calabria is needed in order to validate the ancient accounts. The second section will also describe a typical form of land organization in Southern Italy called massa fundorum. This land management system likely had a great impact on the successful economy of the late Roman elite, including that of Bruttii. In all of this, it is essential to discuss the presence of senatorial families in Bruttii. This may elucidate both the cause and effect behind the general wealth displayed at buildings such as the Villa at Palazzi di Casignana. During the third and fourth century AD the creation of an aristocratic network in Bruttii likely occurred in dynamic urban centers, particularly that of Rhegium, a political administrative capital and important harbor of the Italian peninsula. Its strategic position in the heart of the Mediterranean and its function as the seat of the province’s governorship under Diocletian (284-305 AD) and after his administrative reform of the provinces

171

must have played a pivotal role of the elite in the region and in the economic growth of Bruttii.532 Rhegium became the seat of the corrector of Lucania et Bruttii together with Salernum, which was part of Lucania. Not surprisingly, the greatest wealth is evident in the southern part of Bruttii rather than in the northern during the third and fourth century AD.533 Since the 1960s new archaeological discoveries, the study of pottery’s distribution and inscriptions uncovered in Bruttii have offered valuable insight into the dynamic of the urban and rural occupation of the region during the Late Empire and has helped to identify what aspects the late Roman society of Bruttii shared with other parts of the Mediterranean.534 Further archaeological research in the region will establish new links to the ancient sources, helping to show how the elite impacted the economy in this region and its role in the later history of the Empire.

Section 8.1: the late Roman elite and their economy

Land - the main source of revenue: The accumulation and distribution of aristocratic estates Scholars agree that the huge fortune typically displayed inside later Roman villas across the Empire, such as in the Villa at Palazzi di Casignana, was generated from a rural economy with strong commercial tendencies.535 The aristocratic families of the third and fourth century AD became extremely rich thanks to their investment in land, just as they did during the Late Republican and Early Imperial times.536 Land produced enough income to satisfy the needs of the owners and to enable them to engage in profitable commercial exportation to near and distant markets. Land ownership and exploitation of vast rural estates (latifundi) were the main sources of aristocratic wealth, reaching new heights in many regions of the Empire, including Bruttii during late antiquity. Many latifundi were created through the incorporation of small properties (and their buildings) into the hands of few landowners who increased their assets and kept investing in land that generated more wealth.537 In these latifundi often an embellishment and enlargement of villas occurred. Hyginius Gromaticus (98-117 AD) refers to possessores who acquired estates preserving some villas and abandoning others, confirming the archaeological

532 Diocletian doubled the number of provinces from fifty to almost one hundred in order to prevent local usurpations, facilitate the collection of taxes and supplies, and the enforcement of the law. The provinces were grouped into twelve , each governed by an appointed magistrate called a vicarius, or “deputy of the praetorian prefects.” See BARNES, T.D., The New Empire of Diocletian and Constantine, Cambridge, Mass.1982; REES, R., Diocletian and the Tetrarchy, Edinburgh University Press 2004. 533 SANGINETO 1994. 534 Advancement in the Roman studies of the region is well represented by academic publications that became available to researchers. See LRBC (1965); HAYES 1972; Atlante delle Forme Ceramiche, 1981 (vol. 1); idem 1985 (vol. 2); RIC X (1994); PANELLA 1988. 535 General scholarship on late Roman villas links the wealth of these establishments with land exploitation and commerce led by the Roman upper classes. For Italy, see GIARDINA 1986a and 1986b. 536 Among a vast literature on the topic see CRACCO RUGGINI 1961; PERCIVAL 1976 ; BALMELLE 1992 and 2001; GIARDINA 1986a and 1986b; VERA 1986 and 1988; BROGIOLO- CHAVARRIA 2007. 537 CHRISTIE, N.(ed.), Landscapes of Change: Rural Evolution in Late Antiquity and the Early Middle Ages, 2004, p. 87. 172

evidence.538 Senators derived their fortune from the land, and in fourth century AD they were still substantial landowners.539 Even decuriones, the municipal elite, high military officers and Christian priests in late Roman times owned sizeable estates. Yet, the sheer number of properties accumulated by the Roman aristocrats who often inherited and handed down estates from one generation to the next, could not be easily matched by other social classes.540

The accumulation of land

Aristocratic estates were different in size and some were immense, varying from region to region and being influenced by the geography of each area. In the Naturalis Historia, the Elder Pliny (77 AD) described and deplored the growth of huge estates: ‘…latifundia perdidere Italiam vero et provincias.’541 He argued that the latifundia had ruined Italy and would ruin the Roman provinces as well. In ca.100 AD the Younger Pliny in a letter addressed to Calvisius Rufus wrote about pulchritudo iungendi, referring to the trend of aristocratic landowners who acquired new land to augment or extend their existing properties.542 This phenomenon intensified after the Empire was reorganized under Diocletian (284-305 AD), causing important changes in the administration, economy and society of all the Roman provinces.543 Étienne demonstrated that land accumulation was the principal economic goal of the elite in the fourth- fifth century AD.544 A high concentration of fundi in the hands of very wealthy landlords is a phenomenon described in texts written by the Fathers of the Church, and in the Codex Theodosianus.545 A passage from a sermon written by St. Ambrose, in the last quarter of the fourth century, laments the continuous abuses of some possessores, who arrogantly and greedily redrew the boundaries of their properties often invading the fields of neighbors.546 The poem by Ausonius titled De herediolo was dedicated to a ‘small’ inherited estate located in Aquitania (South-west France) at Gironde.547 The poem provides the most precise information on the size of an aristocratic estate in fourth century AD. The estate included 1050 iugera of cultivated land (ca. 250 hectares) a size that would classify the property among what we would regard as very large estates, the most common size being between 200 and 300 hectares, more rarely 400 hectares.548 Noteworthy is that 1000 iugera are valued at one million

538 Fragments of a work on legal boundaries attributed to Hyginius will be found in LACHMANN, C. F., Gromatici Veteres, I, 1848. See, CUOMO, S., Technology and Culture in Greek and Roman Antiquity, Cambridge- UK, 2007, pp.110-113. 539 CRACCO RUGGINI 1961; ARNHEIM 1972; MATTHEWS 1975; LANÇON 1999; SALZMAN 2002. 540 The process of land accumulation in Italy has been studied by CRACCO RUGGINI 1961, VERA 1986, 1992-93; 1995, 1999a, 1999b; GIARDINA 1986a and 1986b. 541 Pliny, NH, 18, 7.35. 542 Pliny, Epistulae, 3, 19. 543 GAUDEMET, J., “Mutations politques et géographie administrative: l’Empire romaine de Dioclétien (284) à la fine du V siecle”, in La géographie administrative et politique d’Alexandre à Mahomet, Leiden 1981, p.259; GIARDINA 1986a and 1993. 13 ÉTIENNE 1992, pp. 305-311. At this time the terms fundus, praedium, ager, villa, cortis etc. are all synonyms. See FUSTEL DE COULANGES, N.D., “L’Alleu et le domaine rural pendant l’époque mérovingienne,” in Histoire des Institutions Politiques IV, Paris 1889, pp. 2 ff. and p.15. 545 References to both sources are in CRACCO RUGGINI 1961, p. 29. 546 CRACCO RUGGINI 1961, pp. 25-26 with notes. 547 BALMELLE 2001, p.55 with notes; Ausonius, De herediolo, 21-24. See commentary by ÉTIENNE R., 1992, pp. 305-311. GREEN 1991. 548 Du latifundium au latifondo - - 173

sesterces, which is the senatorial census.549 To become or remain a senatorial aristocrat with the rank of clarissimus, it was required by law to make an official declaration (professio) stating the income, the acceptance of senatorial rank, the province and city in which the aristocrat claimed residency.550 Another important source that describes the diffusion of elite landownership is the Life of Melania the Younger written by Gerontius, a fifth century biography of an ultra rich Christian ascetic woman born of prestigious senatorial family.551 This text describes the enormous properties that Melania owned and how widely dispersed her landownership was across the Empire. When Melania fled to North Africa with her husband, she donated to the church of Thagaste, on the Numidian border, an estate that was larger than the town itself.552 The estate was inhabited by numerous artisans who worked in gold, silver and copper.553 This passage also indicates that aristocrats at that time likely derived their wealth not only from agriculture products, but also from different types of industrial manufacturing. Having artisans on the estate was recommended by the late fourth century agronomist Palladius in his treaty Opus Agriculturae. According to Palladius, the presence of craftsmen on the estate would help discourage the peasants from going to town to find essential services they needed. 554This would have favored the villa’s self-sufficiency, and landlords could rely upon the presence of work force that was dependent on the estate. A passage by Olympiodorus recounts how Roman senatorial mansions were like small cities, containing their own hippodrome, fora, temples, fountains and baths.555These texts describe how late Roman private estates were vast and achieved an extreme level of autonomy. Arnheim describes them as ‘self-contained social and economic units.’556 Although the excavations at Palazzi di Casignana have not uncovered farming or storage facilities, it is likely that Villa at Palazzi di Casignana served not only as a retreat residence but was also a productive and administrative center for a wealthy, upper class family, as it was the case for several other late Roman villas. Some of them featured a pars fructuaria, an area devoted to processing agricultural products including oil and wine. The pars fructuaria also provided storage of goods, a shelter for farm animals, and workshops for craft activities. This sector of the villa was usually separated from the pars urbana, the residential area that was more

1992.Published in 1995, Diffusion, De Boccard (Paris). The extension of the estates in the suburbs of Rome was about 50 hectares, see VOLPE 2003, pp.225-226; BROGIOLO-CHAVARRIA 2007, pp.15-16. 549 Du latifundium au latifondo, 1995, pp.459-460. 550 CHASTAGNOL 1992, pp. 298-299; Codex Theodosianus 6.2.13. 551 CLARK, E. A. The Life of Melania the Younger: Introduction, Translation and Commentary, 1984, pp. 18, 21- 22; 21 (L). 552 Vita Melaniae, 10(L), 11, 18-20, 34; Symmachus, III, 53; C.Th. VI, 2, 16 (395). For a discussion on the eversive character of such donations see GIARDINA 1986d and 1994. 553 Vita Melaniae; BROWN, P., Poverty and leadership in the later Roman Empire, Hannover 2002, p.55; WICKHAM, C., Framing the early Middle Ages: Europe and the Mediterranean 400-800, Oxford 2005, pp.163 ff. 554 Palladius I, 6.2: “Ferrarii, lignarii, doliorum cuparumque factores habendi sunt, ne a labore sollemni rusticos causa desiderandae urbis avertat.” 555 OLYMPIODORUS, frg.43. ARNHEIM 1972 p.145. A passage by Agennius Urbiculus in “De Controversiis Agrorum,” describes the huge sizes of North Africa estates and the presence of a fairly large population around the villa in the manner of a city. The text reads: “especially in Africa where private landowners have estates not smaller than the territories of commonwealth; rather the estates are larger than many territories, but the private landowners have a not inconsiderable plebeian population in their estates and villages surrounding the villa like fortifications.” See Corpus Agrimensorum Romanum, (Teubner), p. 45. A treatment on the agriculture and estates of Africa is in KEHOE D., The Economics of Agriculture on Roman Imperial Estates of North Africa, p. 74. 556 ARNHEIM 1972, p.145. 174

or less monumental depending on the owner’s wealth. Desenzano in Northern Italy (fourth century AD),557 Faragola in Puglia (fifth century AD),558 and the villa at San Giovanni di Ruoti in Basilicata (fifth century AD),559 provide examples of villas that combined a productive establishment with a lavish residential area. In Bruttii, the late imperial villa of Naniglio, where wine was produced perhaps with other agricultural products, and the villa at Gioiosa Marina where a pars fructuaria has been identified with a figlina, together with other villas discovered near important Roman towns of Bruttii, such as Vibo Valentia and Thurii-Copia, indicate how important it was even for the Roman aristocracy of this region to combine otium (leisure) with successful negotium (business).560 It is to be hoped that new discoveries at Palazzi di Casignana and elsewhere in Bruttii will expand our picture of the entrepreneurial elite throughout the end of the Roman period confirming that this region afforded the Roman elite similar opportunities of increasing their wealth as did other areas of Italy and elsewhere in the Empire.

The wide distribution of aristocratic estates across the Empire

The luxurious character of Villa at Palazzi di Casignana, and that of many late Roman villas, does not necessarily indicate that having abandoned their urban houses and retired from the public life and affairs, the late Roman owners lived permanently in these residences.561 In fact, it is possible that Villa at Palazzi di Casignana was one of the many estates owned by a wealthy fourth century noble who may have visited this estate only very occasionally. Since early imperial times, the custom for the Roman elite was to possess several estates scattered in different parts of the Empire.562 Pliny (NH 18.7.35) reported that during the reign of Nero six owners possessed half of the province of Africa: “The men of olden times believed that above all moderation should be observed in landholding, for indeed it was their judgment that it was better to sow less and plow more intensively. , too, I see agreed with this view. To confess the truth, the latifundia have ruined Italy, and soon will ruin the provinces as well. Six owners were in possession of half of the province of Africa at the time when the Emperor Nero had them put to death (Loeb Translation by H. Rackham).”563 Roman latifundia grew through the conquest of areas such as northern Africa and south of Spain.564 By the second century AD, the Roman upper classes who owned latifundia in many

557 SCAGLIARINI CORLAITA 1994 and 1997. 558 VOLPE 2004, 2005a, 2005b. 559 SMALL 1980, 1981, 1991, 1998, 2005; SMALL-BUCK 1980, 1994. 560A discussion on this villa is in Chapter 6 of this study. 561 ARCE 1997b, pp.23-24. 562 Seneca (c. 4 BC- 65 AD), in his Epistulae, 87,6 writes: divitem illum putas quia aurea supellex etiam in via sequitur, quia in omnibus provinciis arat (You think that the person who brings with him in his travels even his golden plates is a rich man, and because he farms land in all these provinces). 563Pliny was against the view held by Columella who supported the profit-oriented villas. Pliny’s writings reveal a conservative reaction to the new attitudes of the early imperial upper classes towards increasing their revenues. Ownership of land was the only acceptable source of wealth for senators who delegated their freedmen to commercialize their products in order to silently gain substantial profits without being morally disqualified. See MARTIN, R., Recherches sur les agronomes et leurs conceptions économiques et sociales, Paris, 1971. Pliny's six references to latifundia are in his Naturalis Historia 13.92, 17.192, 18.17, 18.35, 18.261 and 18.296. 564The large villa estates of Campania and around Rome, in the valley of the Po and in southern Gaul were typically self-sufficient establishments. Latifundia could be devoted to cattle rearing or to cultivation of olive oil, grain, and 175

regions had sufficient capital to improve their crops and livestock. They virtually supplanted the small farms, both in Italy and in the provinces, since these could not compete with the high productivity of the large scale estates.565 Landownership consolidated into fewer hands, especially among the senatorial elite. Not only the accumulation but also the wide distribution of estates was a characteristic of the imperial centuries of the Empire. The phenomenon reached a climax in the fourth-fifth century AD. Literary texts recount that many aristocrats owned properties scattered across the Roman world (in , Aquitania, Africa, Sicilia etc.). They resided in their country villas for short periods of time rather than permanently, and visited their villas only rarely, if ever. Landowners generally seem to have owned their fundi in several regions rather than grouped together in one territory. This was so that they could rely upon a more steady income from exploitation of lands that were geographically and agriculturally diverse.566 Nevertheless, there is evidence that members of the late Roman upper classes sometimes owned fundi distributed in just one particular region. This is proven by several papyri from Ravenna or by wills that recorded aristocratic donations to the Church.567 Q. A. Symmachus owned estates in Sicilia and Southern Italy but does not mention in his letters any stopover in these properties. He recalls only one trip to visit his properties in Africa.568 He administrated directly his properties in the Campania-Latium areas because these were close to Rome, where he held important magistracies. Whenever he resided in his country villas, he combined otium (leisure) with negotium (business).569 In fifth century AD, Paulinus of Pella writes that he was never able to visit in Achaia and Epirus some estates that he inherited from his mother.570In the same century, the historian Ammianus Marcellinus deplored the extravagance of Roman senators who owned estates scattered all over the world (patrimonia sparsa per orbem).571 Ammianus portrays Petronius Probus, cousin of Melania the Younger and a leading Roman aristocrat of the later fourth century, as a vain and rapacious man who “owned estates in every part of the empire.” 572 Because of the dissemination of fundi all over the Empire, often the owner would be an absentee landlord who entrusted the management of his estates to short-term tenant farmers known as conductores, or to agents called procuratores or actores (stewards).573 It was

wine. However, in Rome, they did not produce grain and Rome had to import grain. DYSON, S. L., The Roman Countryside, (Duckworth Debates in Archaeology) 2003; MARTIN 1971. 565 On the latifundium stood the villa whose estate was normally run by slaves. In the later centuries of the Empire, slave labor became more expensive, and more coloni, or tenant farmers, who cultivated small plots, gradually replaced them. See below for a discussion on the late antique management of the aristocratic estates. 566 On the estate organization during late antiquity see VERA 1983, 1986, 1994a, 1994b, 1999a; 1999b, 2001;2005; DE MARTINO 1993, especially pp. 810-811; MARCONE 1993, especially pp. 836-839; idem 1997, pp.173-174. 567 BROGIOLO-CHAVARRIA 2007, p.14 with bibliography. For Northern Italy see CRACCO RUGGINI 1961, pp.23-30. 568 VERA 1986, p.260. 569 VERA 1986, pp. 258-259; Symmachus, Epistulae, 1, 5, 2-10; 2, 60; 3, 23; 6, 66; 7, 18. 570 Paulinus of Pella, Eucharisticus, 410-419. The Latin text and its English translation by H. G. Evelyn White, as well as the Introduction, are those found in Volume II of the Loeb Classical Library's Ausonius, pp. 295-351. 571Amminaus Marcellinus, 14, 6, 7-20; 28, 4, 6-27; 27, 11, 1-6. MATTHEWS J., The Roman Empire of Ammianus, Johns Hopkins University Press, 1989. 572 Ammianus Marcellinus, 27,11. 573 ARNHEIM 1972, p. 148. Between the conductor and the coloni who worked the land there were intermediate people who administrated the estates called vilici, actores and procuratores. See MARCONE 1993, p.838. On this complex subject there is a vast literature, see for instance VERA 1986 and 1988; MARCONE 1988 and 1993; DE MARTINO 1993; LO CASCIO 1997; LEWIT 1991; ROSAFIO 2002. 176

considered not appropriate to the dignity of Roman aristocrats to be directly involved in commercial matters. Also for this reason, landowning members of the elite delegated others to administer their estates. The most common practice was to rent out the estates to emphyteuticarii with a long term contract, or to coloni with a short term contract. Peasants also worked the land on behalf of the conductores. Frequently land tenants took advantage of their landlords’ absence to make illegal profits, a phenomenon lamented by Symmachus in his letters.574 A significant reference to disputes over financial administration is also contained in the funerary oration that St. Ambrose wrote for his brother Satyrus who had to travel to Africa to see why rents were not being paid.575 In the De herediolo, Ausonius talks about farmers (cultores agri) without specifying their status. In a letter sent to Paulinus of Nola, Ausonius writes about a man named Philon, who takes care of his estates, referring to him as vilicatus, procurator. Philon seems to be in charge of supervising the harvest of those who work the land.576 Entrusted managers rented and collected the fees on behalf of the estate owners who often lived far away.577 Domini and coloni had shared interests when dealing with their fiscal duties.578 They created a sort of vertical alliance against the fiscal policy of the state. Giardina explains that the land tribute paid by the domini to the state was an important factor that reinforced senatorial dominance and the control of land owners over the rural society within one territory. The possessores typically minimized their own land tribute by having their coloni pay a portion of it as part of the tenancy contract. As a result, these two classes, possessores and coloni, supported one another when dealing with the imperial administration. The coloni found in their domini protection from the imperial taxes and conscription. In one occurrence, when Maximinus Thrax was emperor (235-238 AD), coloni rebelled against the imperial soldiers in solidarity with their domini.579 It is very likely that servi like coloni found refuge and protection in their domini. This would explain why the slaves of Melania and her husband Pinianus refused the freedom granted by their masters and asked instead to remain in the estate where they belonged.580 Melania and Pipianus had possessions in Sicilia that were cultivated by 400 slaves; in the suburbs of Rome they owned lands that were worked by 8000 slaves, confirming a large use of slaves within the late imperial estates.581 Yet, in Late Antiquity slavery, which was typical of the late Republican and early imperial latifundi decreased due to the end of Roman conquests. Based on the available documentation we do not know when exactly and how slavery declined in comparison to the number of coloni.582 Some have connected this reduction with the rising of Christianity. But

574 Symmachus, Epistulae, 1.10, 7.2. 575 St. Ambrose, De excessu fratris Satyri 26; VERA 1986, pp.231-276; SALZMAN 2002, p.52. 576 On this well known texts by Ausonius see GREEN, R.P.H., ed., The Works of Ausonius, Oxford 1991, xvii, p. 219-221); ÉTIENNE 1992a, p.306. 577 ARNHEIM 1972, pp.151 ff. 578 On the economic implications of late antique fiscal legislation and the legal status of coloni see KEHOE 2007. 579 GIARDINA 1986a, p. 29, note 196. 580 GIARDINA , A., “Lavoro e storia sociale: antagonismi e alleanze dall’ellenismo al tardoantico”, Opus, I (1982), pp.128 ff. 581 See The Cambridge Ancient History, vol.13, The Late Empire AD. 337-425, by A. Cameron, P. Garnsey, 1998, p.294. Not all slaves were employed as laborers in the field. 582 VERA 1986, p. 3. 177

this tendency had begun before. As Lançon points out, foreign conquests had finished with Trajan’s acquisition of and with them the renewal of the slave population. Moreover, Christianity at the beginning did not make an issue about the status of slavery, but merely exhorted masters to treat their slaves humanely.583 Slaves gradually improved their conditions and were able to rent a plot of land.584 Generally, slaves were being replaced - at least in some parts of the Empire, by rent-paying freemen who worked the land and shared the profits with landowners becoming increasingly tied to the land.585 In 1889, N.D. Fustel de Coulanges studied the integration of slaves with land-tenants on the Roman villas’ estates.586 He mentioned the case of the Sabine farm that Horace received as a gift from Maecenas. Eight rural slaves lived on this farm and also tenants. According to P. Rosafio this integration dates back to the villa model described by Cato. He also argues that tenancy is present in Varro (I.2.17), although it was Columella who explicitly recognized its importance (I.7.4). Rosafio cites other first century AD literary sources, such as Seneca who in one of his letters (Ep. 123.1-2) tell us about his Alban estate and mentions a vilicus and a colonus. Even Martial in a few epigrams mentions the existence of both slaves and coloni when referring to the villas of two of his contemporaries. (II.11.8-10; III.58.20).587 The Opus Agriculturae by Palladius does not mention slaves, except for one instance (1.6.18). Nevertheless, this does not signify that slaves were not an active work force on the late Roman aristocratic estates. Palladius gives us little information about the manpower organization on the farm in general. As Giardina states, it is possible that Palladius does not mention slaves because at that time it was irrelevant to distinguish between slaves (servi) and land-tenants (coloni). Likely, they both played a similar role on a late Roman latifundus and were linked to the villa system representing a new category of serfs.588 In Bruttii, it is not possible to say how diffused slavery and tenancy were on the late Roman estates based on the textual evidence and current material data. Further archaeological research on the countryside may also help to determine the extension of late Roman latifundi. An idea can be drawn from texts dated between the late fourth and sixth century AD that report private aristocratic donations to the Church.589 It can be assumed that, as for other areas of Italy, slavery decreased substantially although it did not disappear completely. At the end of the sixth century, the letters of Gregory the Great still mention its existence on the Sicilian latifundi.590 Giardina highlights how religion was an important vehicle of aggregation between domini and coloni within the late Roman senatorial properties.591 According to Giardina, this emerges clearly in several letters by St.

583 LANÇON 2000, pp. 83 ff. 584 ROSAFIO, P., “Slaves and Coloni in the Villa System” in Landuse in the Roman Empire, CARLSEN J. (ed.), Analecta Romana Instituti Danici, Suppl. XXII, Rome 1994, pp.145- 158, especially p.145. 585 MARCONE, A., Il colonato tardoantico nella storiografia moderna (da Fustel de Coulanges ai nostri giorni) Como 1988; CAPOGROSSI COLOGNESI 1982 and 2002. 586 FUSTEL DE COULAGES N.D., L’alleu et le domaine rural pendant l’époque mérovingienne, Paris 1889; ROSAFIO 1994, pp.147 ff. 587 ROSAFIO P., “Slaves and Coloni in the Villa System” in Landuse in the Roman Empire, CARLSEN J. (ed.), Analecta Romana Instituti Danici, Suppl. XXII, Rome 1994, pp.145- 158. 588 See GIARDINA 1986b, p.33; idem 1986 c; WEBER, M., Storia economica e sociale dell’antichità,’ Roma 1981; CAPOGROSSI COLOGNESI L., “Grandi proprietari, contadini e coloni nell’Italia Romana (I-III d.C.),” in Società romana e impero tardoantico, vol. I, GIARDINA A. (ed.), Roma 1986a, pp. 325-365. 589 FIACCADORI 1994, p.710 and passim. 590 Gregorius Magnus, Epistulae, I.44, I.70. 591 GIARDINA 1997, 1998, 1999. 178

Augustine. In one letter Augustine praises an ex-proconsul (perhaps of Africa) named Pammachius, who owned estates in Numidia, because he persuaded his coloni to renounce the Donatist heresy.592 In another letter, Augustine writes to Festus, who at that time was occupied in a public service, asking him to send to his estates in the territory of Hippona some of his domestici or amici in order to prevent the emergence of the same heresy.593 The role of religion in establishing coloni inside the estate is also evident in the Life of Melania, with regard to the property that Melania and Pininaus donated to the Church of Thagaste.594 That estate had two bishops, one Catholic and one heretic. (Vita Melaniae, (L), 21, 16, ff.). Their presence was assured by the domini in order to meet two different spiritual needs of their coloni. The care that domini gave to their coloni shows how essential it was to provide important services at the estate. Artisans and bishops represent two elements which contributed to the autonomy of the villa and its estate, encouraging a community to grow and its members to remain close by. At Palazzi di Casignana, it is possible that religion found its space within the main core of the complex as it is discussed in Chapter 9 of this study.

The huge wealth of the late Roman elite

The owner of Villa at Palazzi di Casignana most likely enjoyed a social, political and economic status not much different from that of the ultra rich elite portrayed in the writings of Symmachus (ca. 340 – c. 402), Ausonius (ca. 310–395), and Sidonius Apollinaris (ca.430-489). The immense scale of the estates in late Roman times generated a high volume of income for the elite, stirring awe and disapproval among contemporary writers. Surviving fragments by the fifth century historian Olympiodorus of has allowed scholars to estimate the income of an elite Roman family from landed property was about 4000 gold librae, plus the value of goods in kind, estimated to 1300 librae.595 This represented an income equal to 370,000 solidi,596 a colossal sum, but small in comparison with the total value of the family inheritance of these rich gentes.597 Thus, Olympiodorus provides evidence that land ownership produced income from both the annual rent and sale of agricultural products.598 Since homes were not secure, Roman clarissimi kept their gold librae, whether as ingots, solidi or precious objects, in safes situated in the Trajan forum until mid-fourth century.599 The wealth of Roman nobility may have truly been incalculable, but Ammianus Marcellinus provides us with information regarding the assets of several Roman nobles.600 He reports that

592 See note above. Augustine, Epist., 58, 1(CSEL 34, 2, p. 217). 593 Augustine, Epist., 89, 8 (CSEL 34, 2, p. 424). 594 SPIDLÍK 1996. 595 One Roman libra was equal to 45 aurei. 596 Solidus (plural: solidi) is the standard Roman Imperial gold introduced by in the early fourth century in place of the aureus. The solidus had a value of 1/72nd of a Roman libra. See, The Cambridge Ancient History, Volume 12, The Crisis of Empire, A.D. 193-337 (2nd ed.), by Alan Bowman, Averil Cameron, Peter Garnsey (Hardback, September 2005), p. 335. 597 LANÇON 2000. 598 These sums can be compared to those accumulated by Roman magnates during the late Republic. For instance, Crassus is said to have owned 40 or 50 million denarii (Pliny, NH, 33, 134), whereas 175 million denarii (, Crassus, 2). See JONES A.H.M., The Roman Economy. Studies in Ancient Economic and Administrative History, Oxford 1974, p. 121. The solidus was re-introduced by Constantine in 310. For a discussion on the gold solidus see below in this chapter. 599 LANÇON 2000, pp.63 ff. 179

Quintus Aurelius Symmachus, who according to Olympiodorus led a fairly moderate life style even if he spent 2000 pounds of gold on his son’s praetorian games,601 owned profitable estates in Samnium, Apulia, Sicilia and Mauretania, and for three leisure suburban villas in Latium and 12 country villas in Southern Italy, including six on the Bay of Naples.602 Ammianus describes the elite of his time as moving frequently between their properties. He sarcastically equates their journeys with the marches of .603 Gerontios, the biographer of Melania the Younger, illustrates the immense wealth held by her noble family in fifth century that can be representative of a trend that developed in the previous century. When Melania and her husband Pipianus renounced their assets in order to live an ascetic life, it took seven years just to sell the many properties they owned in Africa. Gerontios explains that the sale of the estates in other provinces left the local buyers with little cash.604 Melania was the daughter of Valerius Publicola, who was prefect of Rome in 375. She belonged to the eminent family of the Valerii, and was connected to the influential families of the Aradii and Britti. Her mother, Albina, was the sister of Caeonius, who had been quaestor of the sacred palace, urban prefect and praetorian prefect. At age fourteen Melania married her distant cousin Pipianus, who was born of two patrician families, the Aradii and the Valerii. Pipianus owned properties in Campania, Sicily, Spain, Africa, Mauretania, Numidia and Britain. Gerontios writes that he heard Pipianus saying that his annual income amounted to 12 myriads of gold (120,000). The enormity of this amount leads one to believe that he was talking of solidi and not librae.605 When Pipianus’ assets were added to the revenue generated by his wife Melania’s properties, the annual income of this ultra rich couple was immense. According to Gerontios, Melania and Pipianus sold their possessions in order to distribute the profits to the poor and Christian churches. They raised an astonishing sum of 3,240,000 solidi, which they distributed to different provinces throughout the Empire. Their assets were so great and so widespread across the Empire that the couple had to seek the intervention of imperial forces to deal with conspiring and greedy people attempting to wrongfully possess their properties. Serena, the niece of the emperor Theodosius, and her cousin Honorius were also Christians. They helped Melania and Pipianus to obtain the mediation of bishops and state supervision to protect the sale of their properties.606 Although the sheer size of Melania and Pipianus patrimony was extraordinary, it was analogous to that of the highest late Roman elite and possibly akin to that of the Villa at Palazzi di Casignana’s owner.

600 Ibidem. 601 OLYMPIODORUS, Frg. 44. 602 Symmachus also owned three houses on the Roman Caelian Hill. For an account of Symmachus’ properties see SEECK, O., (ed.), Symmachi Opera, Monumenta Germaniae Historica, Auctores Antiquissimi VI (Berlin 1883), pp.60-66; D’ARMS, J.G., Romans on the Bay of Naples, 1970, pp.226-229. 603 Ammianus Marcellinus, 28.4.18. 604 Vita Melaniae, 12, 20-21, 34. 605 LANÇON 2000, p.63; ARNHEIM 1972. 606 This episode marks a significant change in Rome’s political and social customs. Until then, the emperors had shown the opposite interest, that of preserving the fortunes of senatorial families, who were the guarantors of public benefaction and costly magistracies. In fact, that Melania and Pipianus divested themselves of all their inheritance was regarded insanity by their peers. The entire system of aristocratic values was subverted by their behavior and choices. Gerontios also tell us that the Melania and Pipianus were the first people of senatorial rank who gave away their assets and the vanities of the world. They abandoned their silken garments much in fashion for the clarissimi at that time to wear cheap clothes made with coarse wool. See LANÇON 2000, p.73. 180

The Impact of a Newly Monetized Economy in the Fourth Century AD

a. By 320 AD a loaf of bread might sell b. The gold solidus was not fixed in value for two silvered nummi like this one of in terns of the copper coins. By the Crispus. This sample is midway in size time of this solidus of Theodosius the and purchasing power for this value could be as much as 7,200 denomination issued first by Diocletian nummi and could buy 30-40 modii of as a large silvered coin and ending in wheat. The daily wage for a the Byzantine period as a tiny scrap of cavalryman was then 180 nummi so a copper. Prices followed a similar path solidus represented the pay for a month with a modius of wheat worth 2 nummi and a half. in 305 selling for 40 nummi in 327 AD. 8.1. The nummus of Crispus and the gold solidus of Theodosius.607

The huge wealth attained by the Roman elite during the fourth century was fueled in large part by the monetary reform carried out by Constantine. The monetary reform was recently analyzed by J. Banaji, who shows the impact that this reform had on the economic development of the fourth century elite, setting the foundation for Late Antique prosperity.608 Banaji draws from inscriptions, written texts and archaeological data concerning North Africa and the East. Banaji’s work can be used as a lens for how the market economy flourished across the Empire, particularly in rural areas through the fourth and fifth centuries. Banaji, following the arguments of the Swedish historian G. Mickwitz,609 maintains that the new gold solidus revolutionized the later Roman economy. It established a secured stable currency and enabled clever entrepreneurs to accumulate fortunes that they subsequently invested in land. According to Banaji, the new currency system produced an expansion of the monetized economy. In 310 Constantine significantly bolstered the gold solidus, which had been first introduced by Diocletian (fig.8.1.). The new solidus weighed 1/72 lb and began to reform the monetary system. A drastic devaluation of the copper and silver coins occurred, which until then constituted the bulk of circulating denominations, causing a progressive increase of prices. .

607 After http://dougsmith.ancients.info/worth.html. 608 BANAJI 2001. 609 MICKWITZ, G., Marketing and Competition. The various Forms of Competition at the Successive Stages of Production and Distribution, Helsingfors, 1959. 181

Mazzarino offers a comparison between the purchasing power during the time of Diocletian versus that of Constantine’s time.610 Under Diocletian, a golden libra (equal to 60 Diocletian aurei or to 10,000-12,000 denarii)611 could purchase 1250-1500 sextarii612 of wine. With Constantine, one golden libra (which at this time was equivalent to 60,000 denarii, or 72 solidi) could buy more than two times that amount, approximately 3600 sextarii of wine. Prior to Constantine’s reform, Diocletian artificially lowered the value of gold, measuring it in denarii, and thus favored the denarius over the solidus, which helped the humiliores (the poor strata of society) who were anchored to the copper currency. Constantine did the opposite, and increased exponentially the value of gold coinage compared to that of copper.613 He abandoned the protection of the copper’s value, greatly reinforced the gold solidus, and by doing so, harmed the needier classes whose purchasing power was based on the silverwashed copper coin. From Constantine onward, and throughout the Byzantine era, prices were based on gold and calculated in relation to the gold solidus rather than copper or silver. In order to prevent the hoarding that continued to damage the monetary economy, Constantine ‘liberalized’ the price of gold and allowed it to rise. This measure naturally advantaged hoarders and the richer classes of society.614 Constantine released more gold solidi into the system, a practice that his successors fostered.615 Constantine’s reform changed the economy of the Empire, causing significant social repercussions. An anonymous author writing after the emperor’s death severely criticized Constantine’s policy, arguing that it ruined the needier classes by switching from copper to the gold-based coinage monetary system. He attributed the rise of wealth among the elite to Constantine’s reform (De rebus bellicis 2.1-3).616 The afflicta paupertas (defenseless poor) widened since the lower classes were attached to the copper currency. The divites (rich classes) became even wealthier than before since they possessed most of the gold and their buying power increased. As a consequence, society became more rigidly hierarchical, from the social and economic points of view. At the peak of the social pyramid were the clarissimi and perfectissimi, the senatorial class, who owned large quantities of gold. In the middle were the civic and military officials. At the bottom of the social ladder stood the large population of poor, struggling for survival. Constantine’s reform stimulated a buying spree among the elite who had purchasing power as their gold was in demand. The emperor Julian and subsequent emperors began lowering prices in attempt to alleviate the social conflicts. Price reductions continued throughout the Byzantine era until the ninth century. One of the measures taken was to diminish the amount of golden solidi in circulation in order to lower the

610 For the following discussion see MAZZARINO 1951. 611 The denarius was a silver coin first struck in 211 BC during the , valued at 10 asses. The denarius continued to be the main coin of the empire until it was replaced by the antoninianus in the middle of the third century. The last issuance for this coin seems to be bronze coins issued by Aurelian between 270 and 275 AD, and in the first years of the reign of Diocletian. For more details, see the article “Denarius” in A Dictionary of Ancient Roman Coins by John R. Melville-Jones (1990). 612 One sextarius is equal approximately to 1.14 pint or .546 liters. 613 In 396 AD one solidus will reach the value of 25 librae of copper. See MAZZARINO 1951, p.112. 614 The Cambridge Ancient History, ed. by John Boardman, Alan K. Bowman, Peter Garnsey, Averil Cameron, Cambridge University Press, 2005. 615 LO CASCIO 1995 with bibliography. 616De Rebus Bellicis was written after the death of Constantine I (337) and before the fall of the (476). It is a fourth or fifth century anonymous work about war machines used by the of the time. See IRELAND, R. I. Anonymi Auctoris De Rebus Bellicis, Bibliotheca Scriptorvm Graecorvm et Romanorvm Tevbneriana, Lipsiae 1984. 182

prices. Yet, prices continued to be measured on gold currency leaving the economic and social inequality between the wealthy and the poor unchanged or little changed. Mazzarino looks at the Historia Augusta to see how adverse the senatorial class appears to be in regards to the control and decrease of prices regulated by the State, which often had to respond to the pressure of the lower strata of society. In fact, the plebs of Rome often exploded in riots.617 Ammianus records that under Valentinian I (emperor from 364 to 375 AD), the Roman people set fire to the house of the ex Praefectus Urbis Avianius Symmachus, Q. A. Symmachus’s father, because he allegedly had said that he would rather use his own wine for quenching lime-kilns than sell it at the price which the people hoped for.618 The elite, having largely invested in rural estates and commercialization of its products, dominated the market. Therefore, the elite were hostile to the imperial edicta that put price caps on food, goods and services. Fixed prices favored soldiers, bureaucrats and people at large rather than the aristocratic interests. The impact of the monetary reform set into motion by Constantine greatly contributed to the wealth that prompted the restoration and embellishment of monumental residences like the Villa at Palazzi di Casignana during the fourth century AD. No matter where the Roman aristocrats lived and what political power they exercised, they reached a level of wealth that significantly surpassed previous generations. Olympiodorus reports that the aristocrats of the West in the late fourth century had income of 1333-2000 librae of gold,619 which means that aristocratic fortunes had raised two or three times from late Republic to the end of first century AD, and five to eight times more from 100 AD to the end of 400.620 What had changed from Cicero’s age to Symmachus’ age is that Cicero and his contemporary senators invested all their wealth in Italy, whereas the latter and the fourth century elite owned properties all over the Mediterranean basin. In addition, the imperial government and the monetary policy of early fourth century stimulated their progressive enrichment and expansion of material wealth well beyond the Italian peninsula. The elite of Bruttii participated in the general prosperity of Southern Italy during fourth century AD, and the following sections examine the many sources of wealth available for them in the region.

617 MAZZARINO 1951. 618 Ammianus Marcellinus, 27, 3, 4. 619 OLYMPIODORUS, Frg. 44. 620 See SCHEIDEL-VON REDEN 2002, p.207. 183

Section 8.2: Bruttii in the Fourth Century AD

The growth of the economy in late Roman Southern Italy and Bruttii

Archaeological data collected in recent years is providing researchers with valuable information for painting a clearer picture of villa architecture in Southern Italy during late Roman times. The expansion of the earlier imperial villa at Palazzi di Casignana into a lavish and monumental structure is part of a trend evident in other fourth century villas of Southern Italy, such as the villas at Faragola,621 Porto Saturo in Puglia;622 Masseria Ciccotti in Basilicata;623 and the three Sicilian villas of Piazza Armerina,624 Patti625 and Tellaro.626 Scholars have generally explained this phenomenon as a consequence of the higher concentration of property in the hands of fewer and very wealthy families who came to absorb the estates of the middle and small landowners decimated by the third century economic crisis.627 Vera’s detailed studies provide useful insights into the southern Italian economy of the late Empire and lead us to a better understanding of the economic situation in Bruttii during the fourth century AD.628 Vera argues that the prosperity reflected in the luxurious fourth and fifth century villas uncovered in the last twenty years, invalidates the old pessimistic view held by Rostovtzev and other scholars, that there was a severe decline during late Roman times in Southern Italy.629 On the contrary, the increased population in the countryside and large-scale building in urban and rural contexts of southern Italian provinces like Lucania et Bruttii, Apulia et Calabria and Sicilia reflect significant economic growth at the end of the third century AD.630 The elite classes living or owning land in Southern Italy established a successful economy in the fourth century, which allowed them to sustain a high standard of living as represented by their sumptuous residences. Evidently Bruttii underwent an economic revitalization between the third and fourth century AD.631 This is not to say that wealth in all Southern Italy was distributed among the different classes of society. It is very likely that the lower strata of society remained in relatively poor conditions while the elite invested and grew substantially wealthier than their predecessors. The economic rise of the whole of Southern Italy started under Diocletian (and later with Constantine), when the number of provinces multiplied and Italy was divided into a dozen small units of the provincial type for taxation purposes.632 Under Diocletian, Italy (Diocesis Italiae)

621 VOLPE 2005b, pp. 26 ff. 622 LATTANZI 1973; DELL’AGLIO 1999, pp.33-41. 623 On the territory and villas of Roman Basilicata (Lucania) see GUALTIERI 1994; GUALTIERI 2003 especially pp.167-175. See also SFAMENI 2006, pp.54-57. 624 GENTILI 1999; CARANDINI-RICCI-de VOS 1982. 625 VOZA 1976-77; VOZA 1980-81; SFAMENI 2006, pp.46-49. 626 VOZA 1976-77; WILSON 1990; SFAMENI 2006, pp.49-52. 627 See SANGINETO 1994, p.579. 628 VERA 2005, pp.23-38. 629 ROSTOVZEFF M.I., Social and Economic History of the Roman Empire, Oxford 1957; LO CASCIO E., STORCHI MARINO A. (eds.), Modalità insediative e strutture agrarie nell’Italia meridionale in età romana, Bari 2001, pp. 5-32; LEPORE 1989, pp.119-121; 116-117. 630 VERA 2005. 631 CRACCO RUGGINI 1961, p.154; NOYÉ 1996, p.102. 632 CHASTAGNOL A., “L’administration du diocèse Italien au Bas-Empire,” in Historia 12, 1963, pp.348 ff.; CARRIÉ 1994, 2001; GRELLE, F., “Ordinamento provinciale e organizzazione locale nell’Italia Meridionale” in 184

was divided in two parts: the Italia annonaria for goods’ provision of the imperial comitatus (the Roman army),633 and the Italia suburbicaria for food provision of Rome.634 At this time, each Italian province of the Italia suburbicaria was required to supply food to Rome as a form of taxation. Payments were made in kind through increments of those goods in which the local agriculture and industrial economy was already strong.635 When in the early fourth century the grain from Egypt was diverted to the new capital of Constantinople rather than Rome, the provision of grain to Rome had to come from other regions of the Empire. Southern Italy, Sicilia and produced grain. Even Bruttii, particularly the northern large and fertile plains of the region, supplied grain to Rome. This production became even more important when Africa underwent a crisis and decreased the provision of grain to Rome at the end of the third century. In the mid-fifth century Rome would receive grain almost exclusively from Southern Italy.636 Starting with the emperor Aurelian (AD 270-275) the annona introduced a daily distribution of bread, olive oil, pork and beef meat, in place of grain. Later, Diocletian reestablished as part of the annona wine at a discounted price.637 All of these products came to Rome mostly from Southern Italy.638 Only olive oil came from Africa where it was produced in large quantities for export all over the Mediterranean, which is confirmed by the wide distribution of African oil amphorae.639 According to the Historia Augusta,640 Aurelian was the first emperor to set up free distributions of pork in Rome (pork meat was preferred by the Romans), although there is evidence of occasional meat distributions under the Severi.641 Lucania was traditionally a good

L’Italia Meridionale in Età Tardoantica. Atti del 38esimo Convegno di Studi sulla Magna Grecia, Taranto 2-6 Ottobre 1998, pp.115-139. 633 The Italia annonaria with capital Mediolanum (Milan) was formed by Venetia et Histria; Aemilia et Liguria; Flaminia et Picenum; Raetia; Alpes Cottiae. 634 The Italia suburbicaria ‘under the government of Rome, was composed of Tuscia et Umbria; Valeria; Campania et Samnium; Apulia et Calabria; Lucania et Brutti, Sicilia ; Sardinia et Corsica. See JONES A.H.M., The Later Roman Empire, pp.284-602. A Social, Economic and Administrative Survey, Oxford 1980, pp.824-834; RICKMAN G.E., The Corn Supply of Ancient Rome, Oxford 1980, pp.14-16. 635 On the Late Roman grain supply there is a vast literature. Among the most general surveys see CAMERON, A. The Mediterranean World in Late Antiquity AD 395-600. London 1993, especially Chapter 4, 7; BARNISH, S. J. B. “Pigs, plebeians and potentes: Rome's economic hinterland, c. 350-600 AD”, Papers of the British School at Rome, 55 (1987), pp.137-185; CASSON, L., “The role of the state in Rome's grain trade,” in D'ARMS, J. H. and KOPFF, E. C. (ed.), The seaborne commerce of ancient Rome: studies in archaeology and history. Memoirs of the American Academy in Rome 36 (1980), pp. 21-33. GARNSEY, P. - "Grain for Rome", in GARNSEY P. et alii, Trade in the ancient economy, London 1983, pp.118-130. GARNSEY, P. “Famine in the ancient Mediterranean,” History Today 36.5 (1986), pp. 24-30. Idem, “Famine and food supply in the Graeco-Roman world: responses to risk and crisis,” 1988; GARNSEY, P. AND SALLER, R. P., “The Roman empire: economy, society and culture,” 1987. GARNSEY P. AND WHITTAKER C. R. (eds.), “Trade and famine in classical antiquity,” 1983. 636 CRACCO RUGGINI 1961, p.175, foot note 524 with sources. 637 CHASTAGNOL 1960; DURLIAT 1990. 638 CHASTAGNOL 1960, pp.58-59; 321-330; JONES, pp. 696 ff.; VERA 2000, p. 629. 639 PANELLA, C.-TCHERNIA, A., “Agricultural Products Transported in Amphorae: Oil and Wine,” in The Ancient Economy, SCHEIDEL- VON REDEN (eds), 2002, pp.173 ff. 640 The Historia Augusta is a late Roman collection of biographies of the Roman Emperors of the period between 117 and 284 AD. It is a compilation of works by six different authors (known as the Scriptores Historiae Augustae), written during the reign of Diocletian and Constantine I, but the true authorship of the work, its actual date, and its purpose have long been matters for controversy. BAYNES N.H., The Historia Augusta. Its Date and Purpose, Oxford 1926. SYME, R., Ammianus and the Historia Augusta, Oxford 1968. Idem, Emperors and Biography , Oxford 1971, and Historia Augusta Paper, Oxford 1983. 641 LANÇON 2000, pp.79 ff. 185

producer of pork meat, so it was required to provide it to Rome as a form of taxation. In 452, Lucania et Bruttii paid 6,400 solidi for the annona of pork meat as aderatio to the suarii who were entrusted with provisions to Rome.642 Bruttii and Sicilia had specialized in wine production since the third century, so they were required to furnish Rome with wine in kind (vina fiscalia). Apulia (modern Puglia), was a region where transhumance farming and cattle sheep rearing were practiced, so it produced wool to furnish it to Rome. Apulia also produced grain, as did Campania, Sicilia and Sardinia. Rome stimulated each province to produce the goods that were typical of its economy. Some regions continued to develop certain products motivated by the needs of the state. Others modified their agricultural economy and, at times, they even changed their geographical boundaries in order to participate more effectively in the larger fiscal system. For instance, Metapontum was a center that produced frumentum (grain), and its border was transferred from the province of Lucania et Bruttii to that of Apulia et Calabria for taxation purposes.643 In contrast to what was happening in northern Italy, the annona on vineyards and oxen, was not heavy in Bruttii at this time.644 This allowed the southern Italian landowners to take their surplus and sell it in the open markets of Italy and abroad.645 L. Cracco Ruggini maintains that fourth century imperial fiscal policy was of great advantage to the growth of Bruttii’s economy. In fact, there is evidence that in 323 Bruttii was exempted from taxation on the imperial estates that were rented to long term-land tenants in emphyteusis. This tax immunity was abolished in 383 then reinstated in 408 and, at this time, it was also extended to senatorial property.646 Rich senatorial families actively involved in agricultural business likely saw this tax immunity as an incentive to multiply their productivity, which further increased revenues generated by their properties and commercialization of goods. It is under these circumstances that Southern Italy together with Sicilia and Sardinia strengthened and expanded their economies.647 The late Roman state distributed frumentationes for free to the poor, namely to 180.000 beneficiaries.648 The rest of Rome’s population, which in fourth century AD amounted to about one million people, obtained grain through free market commerce with Southern Italian provinces.649 Vera even highlights that the economy during that

642 GRELLE, F. and VOLPE, G., “Aspetti della geografia amministrativa ed economica della Calabria in età tardoantica, in Epigrafia e territorio, politica e società: temi di antichità romane, ed. by PANI, M., Volume 4, 1996, pp.145 ff.; GRELLE 1998; JONES, A.H.M.,The Later Roman Empire, 284-602: A Social Economic and Administrative Survey, Oxford 1964, p.703. 643 GRELLE, F., in L’Italia Meridionale nell’etá tardoantica. Atti del 38esimo convegno di studi sulla Magna Grecia, Taranto 2-6 Ottobre 1998, p.120 note 13. While Lucania et Bruttii lost Metapontum they acquired the district north of river Sele, which included the Lucanian city of Salernum. Rhegium and Salernum became the two main opposite poles along the coastal route attested by the Iter Siculum of Lucilius (Lucilius 3.124,126,127 M= 15, 16, 17) and later by Cicero (Cicero, Ad Atticum 16.6.1; 16.7.5). See GIVIGLIANO, G.P., “Percorsi e strade,” in Storia della Calabria Antica. Età Italica e Romana, II, ed. by S. Settis, Reggio Calabria 1994, p. 293. 644 ARSLAN, E.A., in L’Italia Meridionale nell’età tardoantica. Atti del 38esimo convegno di studi sulla Magna Grecia, Taranto 2-6 Ottobre 1998, p.404. 645 CRACCO RUGGINI, L., Economia e società nell’ “Italia annonaria.” Rapporti fra agricoltura e commercio dal IV al VI secolo d.C., Milano, 1961, Collana della fondazione Guglielmo Castelli, 30, pp.49-50; idem, “Vicende rurali dall’Italia antica tetrarchica ai Longobardi,” in Rivista Storica Italiana, 76, 1964, pp. 261-286. 646CRACCO RUGGINI 1961, p. 43 and 56; ARSLAN 1998, p.406. 647 VERA 2004, pp.23-38. 648 MAZZARINO 1951, pp. 217-247; CHASTAGNOL, A., La Préfecture urbaine à Rome sous le Bas-Empire, Paris, 1960, p.314; DURLIAT, J., De la ville antique à la ville byzantine. Le problème des subsistances, Rome 1990, pp.110-123. 649 VERA 2005. 186

time called for ‘less state and more market.’650This greatly favored Roman senators who owned huge estates in Southern Italy. They could assert strong influence on the Roman market. According to Vera, it was this strong commercial relationship that Southern Italy had with Rome and its market that determined the fourth century economic rise of the whole of Southern Italy. The fact that food supply to Rome came from Southern Italy more by means of free market than from taxation would explain the economic and agricultural growth of provinces such as Apulia, Lucania et Bruttii, Sicilia and Sardinia. Indeed archaeological data, especially pottery, and literary sources show that the most significant source of prosperity came to southern Italy from its trade specifically with Rome, confirming Vera’s arguments.651 The surplus of grain from Apulia, Lucania and Sicilia, and the surplus of other products particular to each southern region such as wine from Bruttii and Sicilia, and pork from Lucania, supplied Rome and its large population in the fourth century AD. The enterprising elite class of land owners and merchants found in Rome a valuable market for their products, as proven by the diffusion of amphorae imported from Southern Italy.652

Wine and amphorae: production and exportation during the late Roman centuries

A specialization that certainly gave a particular shine to the prosperity of Bruttii in the fourth century AD was the production of wine. Viticulture had a long tradition in the region. It began in Hellenistic times and lasted until Late Antiquity and beyond.653 Wine amphorae produced in the region date from the mid-second century BC to the first century AD. These amphorae were marketed locally and in small quantities elsewhere. 654 Strabo (58-21ca. BC) and Pliny the Elder (AD 23-79) write about the wine of Thurii in northern Bruttii.655 Pliny also talks about the wine of Consentia and Tempsa that were becoming highly esteemed during his time.656

650 Ibidem. 651 Pottery is a significant indicator of commerce, trade, and daily customs for any given period of the Roman past. It often compensates the lack or scarceness of information of the written sources. 652 NOYÉ, G., 1994, p.706; VERA 2004. 653 SANGINETO 1994. 654 DE CARO 1985. 655 Pliny, NH 14, 3, 39 and XIV, 8, 89; Strabo VI,1,14. 656 Pliny, NH 14, 69. See SANGINETO 1994, p.564. 187

8.2. Keay 52 amphorae.657

An exponential growth of local amphorae production occurred between the end of the third and the beginning of the fourth century AD.658 According to P. Arthur, there was a type of wine amphora (Keay 52, fig.8.2) that was produced mainly in Bruttii.659 This is a small, flat- bottomed amphora with a capacity between 12 and 14 liters. The large number of amphorae found in Bruttii point to a remarkable increase of wine production in the region during the late imperial times.660 Arthur’s theory was confirmed by the discovery of Keay 52 amphorae kilns near Rhegium, in Pellaro and at the nearby town of Lazzaro, which are dated between the fourth and fifth century AD.661 With the exception of the kiln in Pellaro, all the amphorae kilns that have been uncovered are located in proximity of an imperial villa, pointing to a central role of villas in the rural and industrial economy of the region. Some of the Keay 52 kilns in southern Bruttii seem to stop in 500, although wine from the region certainly remained available along the southern Italian coast and in Rome.662 The Villa at Palazzi di Casignana has revealed numerous Keay 52 amphorae in the service gallery of the baths praefurnia. They are dated to fourth-fifth century and outnumber the

657 After Guida, p.96, fig.143-144. 658 SCHEIDEL-VON REDEN (eds.), 2002, p.185. 659 ARTHUR, P., “Some observations on the economy of Bruttium under the later Roman Empire,” JRA 2, 1989, pp. 132-142. 660 On the topic see PANELLA, C., “Merci e scambi nel Mediterraneo tardoantico,” in Storia di Roma, 3.2. I luoghi e le culture,” Torino 1993, pp.613-697; PANELLA, C.,- TCHERNIA, A., “Produits agricoles transportés en amphores. L’huile et surtout le vin,” in L’Italie d’Auguste à Dioclétien, Roma 1994, pp. 145-165. 661 ANDRONICO, E., Il sito archeologico di Pellaro (frazione di Reggio Calabria), in La Calabre, pp.731-736; GASPARETT,I G.-DI GIOVANNI, V., “Precisazioni sui contenitori calabresi della tarda antichità (le anfore tipo Keay LII),” in La Calabre, op. cit., pp. 876-979. 662 WICKHAM, C., Framing the Early Middle Ages: Europe and the Mediterranean 400-800, Oxford 2005, p.730. 188

imported amphorae by seven to one.663 The villa is situated in the center of a rural and fertile land that still today yields a good quality wine called Vino Greco Mantonico. It is very possible that farming and wine-making took place on the estate in fourth century AD and that the profit gained from wine making helped build the ostentatious villa. The massive walls discovered underneath the mouth of the nearby river Bonamico might have belonged to a harbor. Its presence would be highly significant especially for the capacity of the villa to interact on its own with overseas markets. The hypothesis that the villa may have contained a Christian space (as discussed in Chapters 9 and 10), and was owned by a Christian dominus is not invalidated by the possible presence of wine production. Early Christian foundations were notably involved in wine-making and gained profits from it, as it was the case at the monastery founded by Cassiodorus at Scolacium in Bruttii.664 Significantly, Keay 52 amphorae have been unearthed in many archaeological contexts throughout the Mediterranean, including Greece, Southern Gaul, Spain and Italy. They range in date from the fourth to the eighth century AD, indicating a continuous wine production and trade undertaken by Bruttii on a vast area and over many centuries.665 In Rome and Ostia, Keay 52 amphorae remarkably represent 60% of the imported wine amphorae. In Rome, a deposit of Keay 52 wine amphorae, dated between 312 and 315 AD, was found on the Palatine, indicating a heavy import of wine from Bruttii.666 The fourth century landowners of Bruttii likely took advantage of the decrease in wine production that affected Northern Italy.667 Pliny the Elder writes that viticulture was the main economic resource of Cisalpine Gaul.668 In fact, Northern Italy had been the traditional wine provider for Rome (vina fiscalia).669 The situation must have changed during the later imperial centuries. Fourth-fifth century sources bear witness to a fall in quality and production of northern Italian wine.670 Ennodius, for instance, deplored the low quality of wines from the Po Valley.671 Vina fiscalia, the wine given to Rome from the provinces as a form of taxation was distributed to the plebs of Rome not for free but at discounted price. The Historia Augusta reports Aurelian’s intention to distribute wine gratis to the plebs of Rome, but this was never enacted.672Another law of 365 AD by Valentinian established that wine would not be distributed to plebs of Rome for free, but instead at a ¼ discounted price from the current market price.673 The possessores of Bruttii, even if they were furnishing wine to Rome as a form of taxation must still have generated large profits from commercializing wine because production was abundant.

663 BARELLO F.-CARDOSA M., “Palazzi di Casignana Palazzi,” in MEFRM 103, 1991, 2, pp.669-687, especially p. 676. 664 Variae 12.12. 665 NOYE 1998, p. 445; ARTHUR, P., “Some observations on the economy of Bruttii under the later roman empire,” in JRA, 2, 1989, pp.132-142; CIARROCHI ET AL., “Produzione e circolazione di ceramiche tardoantiche ed altomedievali ad Ostia e Porto,” in La storia economica di Roma nell’altomedioevo alla luce dei recenti scavi archeologici (Roma, 1992), in DELOGU P.- PAROLI L. (ed.), Florence 1993, pp. 203-246. 666 PANELLA, C.-TCHERNIA, A., “Agricultural Products Transported in Amphorae: Oil and Wine,” in The Ancient Economy, SCHEIDEL- VON REDEN (eds), 2002, pp.173 ff. 667 CRACCO RUGGINI 1961, pp.35 ff. 668 Pliny, NH, 13, 127. 669 On wine provision to Rome from the Italia Annonaria see CRACCO RUGGINI 1961, pp. 35 ff. 670 CRACCO RUGGINI 1961, p.180. 671 Ennodius, Opp. CCCLXV and CCCLXVII; CRACCO RUGGINI 1961, pp. 180 and 535. 672 CRACCO RUGGINI 1961, pp. 38-39. Sextus Aurelius Victor (ca. 320-ca. 390) was the author of a History of Rome from Augustus to Julian published ca. 361. 673 CRACCO RUGGINI 1961; Codex Theodosianus XI 2,2. 189

Written sources show that Bruttii had a notable wine production in fourth century. The Codex Theodosianus (14.4.4) reports that vina fiscalia arrived in Rome from Lucania et Bruttii in 367 AD. The Expositio totius mundi, an anonymous text of mid-fourth century AD, praises Bruttii for being a producer and exporter of vinum multum et optimum. This shows that the production and exportation of quality wine from Bruttii was impacting the economic growth of fourth century AD.674The statesman Cassiodorus (c. 485 - c. 585), a native of Scolacium on the Ionian coast of Bruttii and an influential landowner and governor of Lucania et Bruttii from 515 to 523 AD, described the wines of Bruttii in his letters, praising their high quality and affirming that they were much admired even at court.675 Cassiodorus described one kind of wine produced in Bruttiii as “moderately thick, strong and brisk, of conspicuous whiteness, distinguished by the fine of which a pleasant after-taste is perceived by the drinker.’ He also pointed to the wine’s curative properties: “It constrains loosened bowels, dries up moist wounds, and refreshes the weary breast.” 676 Wine was consumed on a daily basis not only by wealthy Romans, but also by soldiers and by the lower strata of society. St. Ambrose and Sidonius Apollinaris refer to the large amount of wine consumed by the poor.677 Ambrose even shows disapproval at the fact that viticulture is preferred to the cultivation of cereals, which he considers more essential than wine.678 The provision of wine for the population of Rome was important to the emperors. Ammianus Marcellinus reports that numerous riots broke out in Rome over the lack of wine (ob inopiam vini) between 353 and 355 AD, when Symmachus’ father in law, Memmius Vitrasius Orfitus Honorius, was praefectus Urbi.679 Other insurrections occurred after 365 AD during which the house of Symmachus’ father, Lucius Avianius Symmachus, was set on fire in protest against the lack of wine provisions for the people.680 At the end of 365 AD, Valentinian I enforced the exaction in kind of wine from the provinces. This annulled an earlier law by Constantius (354 AD), which established that all the landowners of Italy could provide wine by aderatio (conversion in money of taxes that the State demands in kind), in accordance with a quantity of money (pecuniae quantitas) determined by the praefectus praetorio Italiae Vulcacius Rufinus.681 This guarantee of wine provisions resulted in a gradual price increase of wine throughout the fourth century. There is evidence that between the years 356 and 363 the cost of wine multiplied eight times since the turn of the century. In 301 AD it was possible to purchase 260 wine amphorae with one golden libra. That same golden libra could only purchase 32 amphorae between 356 and 363 AD.682 This eightfold increase in

674 Expositio Totius Mundi et gentium, in J. ROUGE, Paris 1966 (Sources Chrétiennes, 124), p.190; for the production of wine in sixth century see Variae, 8, 31; XII, 4; XIV. A famous passage of the Codex Theodosiani shows that cattle rearing and viticulture were linked (Theodosiani libri XVI, in T. MOMMSEN- P. M. MEYER, Berlin 1905, 14, 4, 4). 675Cassiodorus, Variae, 12.12 and 14. See “Trade and Transfer,” in The ancient economy, SCHEIDEL-VON REDEN (eds.) 2002, p.185. 676 Translation from HODGKIN, T., The Letters of Cassiodorus, 1886. 677 For a commentary on both passages see CRACCO RUGGINI 1961, pp.53 ff. 678 AMBROSE, De Noe, 109; CRACCO RUGGINI 1961, p.180. 679 Ammianus Marcellinus, 14, 6,1. 680 Ammianus Marcellinus, 27, 3.4. 681 Cfr. Codex Theodosianus 11, 2, 2.; CRACCO RUGGINI 1961, p.55, note 26. As pointed out by Cracco Ruggini, this law was established in order to avoid the speculations of tax officials who arbitrarily decided the monetary value of wine. 682 For the increase of wine price see CRACCO RUGGINI 1961, p.35; NOYÉ 1994, p.669. 190

wine prices must have benefited the landowners of Bruttii, especially along the favorable fertile and mild hills of the Ionian and Tyrrhenian coasts. Wine and amphorae making in Bruttii is well supported by ancient sources and recent archaeological findings. Bruttii built a respectable reputation for its production and distribution of wine across the Empire. The abundance of Keay 52 amphorae on the site of the Villa at Palazzi di Casignana strongly suggests that wine was produced on the property. This would frame the villa as a rich residence and possibly also as a hub for wine-making. The opulence displayed by the late imperial villas of Palazzi di Casignana and by those located near Locri (see Chapter 6), may have been the result of local wealth, which very likely was largely based on wine making.

The geographical location of Bruttii

Bruttii benefited from its favorable geographical characteristics and location at the center of the Mediterranean Sea. Rome was reached through the main ports of Thurii-Copia, Vibo Valentia, Croton, Scolacium, Locri, and Rhegium.683 The same ports put the region in contact with the main commercial routes of the Mediterranean. A series of smaller ports have been detected all over the region, indicating there was intense maritime traffic.684 The vicinity of Bruttii to Northern Africa facilitated trade between these two areas. Wine amphorae, both local and imported from Africa, were found in large number in Rhegium.685 From the third century onwards it is Africa Proconsularis with its main products (oil, grain and wine) that will dominate the market of the Italian peninsula and many provinces of the Empire.686 Because of this role of Africa, the economic axis moved towards the Southern part of the Empire, causing a separation of the Mediterranean market from that of the European continent.687 The abundant presence of African ceramics in Bruttii ranging from the fourth until eighth century AD points to intense trade with Africa, most likely through the mediation of Sicily, especially along the coasts. 688 At the Villa of Palazzi di Casignana, there is a significant presence, throughout the final phases of the villa, of African sigillata fragments (D1 type and Hayes 59), which are dated between 320-420 AD. Later amphorae type Hayes 61A with its variants were also found on the site.689 The most recent type of African sigillata found in the

683 For an account of these important Roman ports of Bruttii see SANGINETO 1994, pp.565-567; , . 684 GIVIGLIANO G.P., “L’organizzazione del territorio,” in Annali della Scuola Normale di Pisa, s.3,19,1989, pp.737-764. LA TORRE F., 1999, pp. 28-31. 685 Variae, XII, 14; NOYÈ 1998, p.448 686 CARANDINI A., “Produzione Agricola e produzione ceramica nell’Africa di età imperiale”, in StudMisc, XV (1970), pp. 95-119; Trade in the Ancient Economy, edited by GARNSEY, HOPKINS, WHITTAKER, London 1983, pp. 45-62. 687 PANELLA 1988, pp.618-619; CARANDINI A., “Il mondo della tarda antichità visto attraverso le merci”, in Società romana e impero tardo antico, III. Le merci. Gli insediamenti. Edited by Giardina A., Roma-Bari 1986, pp.4-5. 688 For an account on African ceramics and amphorae imported in Bruttii, see “Atti del Convegno dell’Ecole Française of Rome”, in Melanges Ecole Française of Rome (M), 100, 1991, 1. A large number of African ceramics and amphorae discovered in Scolacium are discussed in RACHELI A., “Le ceramiche da mensa, le anfore”,and in SPADEA R. (ed.), Da Skylletion a Scolacium, pp.147-163. 689 African sigillata is clay tableware with a red-orange coloring and shiny finish produced in the Roman provinces of north Africa from the end of the first century AD and exported in the whole Mediterranean area until Late Antiquity. 191

villa is E Hayes 70, dated to 450 AD, identifying with the last phase of the villa’s inhabitation. 690 Fragments of imported amphorae from the pars orientalis of the Empire dated between 400 and 650 AD are also attested at Palazzi di Casignana, but in smaller quantities. The imported ceramics from both Africa and the East indicate that the Villa at Palazzi di Casignana was inserted in the international circuit of commerce and trade that the fourth century Roman elite of Bruttii successfully administered and promoted. The presence of massive walls underneath the mouth of the nearby river Bonamico might have belonged to a harbour, which would have put the villa in communication also with the hinterland and in contact with overseas market. But further excavations need to confirm this hypothesis.691

Other Resources in Bruttii

Sources and archaeological evidence indicate that the fourth century economy of Bruttii appears to have benefited gradually from a larger utilization of forests, reclaiming of marshes, expansion of pasture-land, the rearing of oxen, pigs, horses, fish farming and mining.692 The most profitable activities in Bruttii are linked to agricultural production, especially viticulture, from the third century AD onward. Pliny the Elder mentions the apple trees of the Consentia area, which produce three times the amount of apples of normal trees. According to Varro it was two times.693Both assessments are indication for how fertile was the land in Bruttii. According to Cassiodorus, late antique landlords in the region drew a good portion of their fortune from horse rearing, combined with the production of milk, cheese, meat and lard.694 The late Roman entrepreneurial elite of Bruttii likely leveraged the 800 km of continuous coastline, deriving wealth from fish farming, fish processing and the production of garum, a costly fish sauce that the Romans habitually used for cooking and also for medical purposes.695 Fish rearing was celebrated at the Villa at Briatico on the Tyrrhenian coast dated to the first century AD.696 The villa contained an exquisite polychrome mosaic representing fishing amorini (fig.8.3).697

690BARELLO F.-CARDOSA M., “Palazzi di Casignana Palazzi,” in MEFRM 103, 1991, 2, pp.669-687, especially pp.673-674. 691 See Guida, p.35. 692 CRACCO RUGGINI 1961, p.154; ARSLAN 1998, p. 405; NOYE 1994, 2002. A negative view of the Roman occupation of Bruttii is held by PLACANICA 1999. 693 Pliny, NH, 16, 115; Varro, De re rustica, I, 7,6; SANGINETO 1994, p.564. 694 CASSIODORUS, Variae, 8.31; NOYÈ, GH., Les Bruttii au VIe siècle, op. cit., pp. 506-507. 695 HIGGHINBOTHAM, J.A., Piscinae: artificial fishponds in Roman Italy, University of Northern Carolina Press 1997. 696ACCARDO 2000, p.181 with bibliography; fish-farming and fish-processing have been well documented along the south and west coasts of Spain and Portugal. See HIGGINBOTHAM 1997, p. 697 The mosaic was removed from the site and relocated on an outdoor panel along the driveway of the Archaeological Museum of Vibo Valentia. 192

3. 8.3.Mosaic of the Villa at Briatico (1st c. AD).698

The villa features four tubs dug in the sea-rocks where fish were reared. The site connected to a small harbor that allowed for trade and commercialization of the products. A harbor was a frequent feature in villas where fishponds were installed. Structures associated with fish-rearing have been uncovered at the train station of modern Reggio (Rhegium), dated to the sixth and seventh century AD.699 Many piscinae and establishments for fish processing may still lie undiscovered underground or sunken along the shorelines of the Ionian and Tyrrhenian coasts of Bruttii. It is possible that fish rearing occurred also at the Villa at Palazzi di Casignana, which is located only 300 meters from the sea. A series of tanks located behind the monumental nymphaeum certainly encourage us to consider this possibility.

698 Photo by author. 699 SPADEA R., “Lo scavo della Stazione Lido (Reggio Calabria),” MEFRM, 103, 2, pp. 689-707. 193

8.4. Tanks behind the nymphaeum at Palazzi di Casignana.700

8.5. Fishpond at Pompei.701

Behind the monumental nymphaeum there are five large rectangular cisterns (fig.8.4). The first one (71) is placed immediately behind the apse of the nymphaeum and is orientated east-west.702 Behind Tank 71 there are other four cisterns (72-75), which are coated with the same hydraulic plaster as Tank 71 and connected with one another by means of quadrangular openings. Tank 71

700 After Guida, p. 63, fig.77-79. 701 After HIGGHINBOTHAM 1997. 702 Tank 71 measures 8.70 m. x 4.90 m. Its interior is coated with hydraulic plaster as well as the bottom surface which is inserted with pebbles. No pipes have been found, but the entire tank slightly slopes towards the nymphaeum to allow water to flow. 194

at some point in time had its opening walled so that it did not communicate with Tank 72 placed behind. Except for tank 72, all the tanks were sunk in the ground, a device that allowed the water to remain cold even during the warm seasons. Several features suggest that these cisterns, especially 72, 73, 74 and 75 placed behind 71, might have been used for rearing different species of fish. The lack of pipes, the closing of the internal opening inside Tank 71 that originally connected this tank to 72; the internal dividers that formed the four tanks 72-75 which were left communicating by means of the openings, and the construction of an external wall around all four tanks 72-75 (perhaps to prevent fish jumping out of the tanks),703 are all changes that might have been determined by the choice of rearing different species of fish in separate tubs on the estate.704 Tank 71 seems to have been more closely connected to the fountain than the other four tanks behind it. It likely furnished water to the monumental nymphaeum. The four tanks (72-75), instead, are similar to one another for plan, size, and for the fact that, contrary to Tank 71, they are sunk in the ground. Their function seems to have been the same but separate from that of Tank 71. All five tanks could contain 193.410 liters of water.705 Yet, lack of water must have not been an issue at Palazzi di Casignana if the owner resolved to build two sets of baths on the estate that contained comfortable tubs and a huge one with a capacity of 126,000 liters (Tub 25). Although it is very likely that all of these tanks were built to supply water for the baths, the daily needs of the villa’s inhabitants, and for agriculture purposes, as suggested by the archaeologists, it is not excluded that they were used also to breed fish, because of the features illustrated above. Leaving this possibility open, future discovery at Palazzi di Casignana will help to clarify the function of these well constructed tanks and what motivated their installment. If they truly were used for pisciculture, the garden, which was already embellished by the monumental nymphaeum, must have been even more astonishing to the eyes of the Roman visitor for the fact that it displayed different kinds of fish. The practice of raising fish existed throughout the Roman world since the later Roman Republic.706 Italy possesses the greatest number of surviving piscinae. In the majority of cases, the piscinae are associated with villas or other private dwellings with easy access to major political and commercial centers by sea or by land. Many fishponds were built close to port facilities or major roads. 707 Pisciculture is included in the manuals of Varro and Columella, indicating that the noble farmer often engaged in this practice. The possession of an artificial fishpond was a mark of high social status. Varro has one of his characters in his dialogue on agriculture state that saltwater fishponds are constructed, stocked and maintained at high cost. The appeal of piscinae salsae, seaside fishponds, is to the eye rather than the purse, which is emptied rather than filled by them.708 Fish and saltwater species in particular figured prominently in the diet of the upper classes. The market value of fish was mostly driven by the demand from

703 The hypothesis of the villa’s excavators, instead, is that this fencing wall may have supported a roofing structure. See Guida, p.63. No explanation is offered in regards to the choice of dividing the tanks in smaller compartments. 704 This possibility needs to be investigated further by experts in the field of Roman pisciculture. For a detailed treatment on Roman pisciculture in freshwater ponds and tanks, see HIGGINBOTHAM 1997, pp.12 ff. 705 Guida, p.63. 706 An artificial fishpond was designed to control the movement of fish, facilitate the flow of water, and in some cases, regulate the degree of salinity within the pond. These structures varied from elaborate sea-girt ponds that covered acres to small freshwater ponds that decorated gardens and peristyles of villas. See HIGGINBOTHAM 1997. 707 Ibidem, p. 240 note 81. 708 Varro, III.17.2: “For in the first place they are built at great cost, in the second place they are stocked at great cost, and in the third place they are kept up at great cost.” 195

aristocratic banquets. Juvenal writes that it was necessary to import fish to meet Italian needs, suggesting a high demand of fish that Italy could not fully satisfy.709 There was great profit in the production of preserved fish and processed fish products as garum, liquamen, allec, muria which were produced in several Italian sites, not only in Gallia and Spain. In Bruttii, Thurii/Copia is singled out for producing garum. A passage by The Elder Pliny (NH 31, 94) praises the town as being one of the best producers of garum other than Pompei, Velia and Beneventum.710 In addition to literary evidence, the production of garum in Bruttii is also suggested by the discovery of locally produced amphorae that feature a wide opening. They likely were used to transport garum.711 Fish-breading may have been a good source of income for the aristocrats over the centuries. Even Cassiodorus in sixth century AD reared fish in his monastery at Scolacium on the Ionian Sea, which took the name ‘Vivarium’ (a place for living things) right from the fishponds he created by the sea.712

709 Ibidem. 710 Ibidem p. 57 note 7. 711 SANGINETO, “Trasformazioni.”, pp.213-214. 712 Cassiodorus, Varie, XII, 14, 4: His victualibus, si nosse, regio illa fecunda est: nam et marinis deliciis copiosa iucunditate perfruitur, quia ibi mare supernum atque infernum, insertis frontibus adunatum, delicias utriusque pelagi in unam congregationem sinus sui volubilitate perducit. Necesse est enim illic et pisces properare, ubi constat et undam posse defluere.

196

8.6. From an eighth-century Bamberg manuscript: the Vivariensis monastery of Cassiodorus. Notice the fish in the vivaria at the bottom of the image.713

The forests of Bruttii were likely an important source of wealth, especially the conifers of the Sila Mountains, from which timber and pitch were extracted in abundance.714 The exploitation of the forest dates back to the Republican age.715 Wood was likely a source of revenue even in the fourth century AD and it was plentiful in the large mountainous areas of Bruttii. In fact, the trees of Sila furnished wood for repairing the roof of St. Peter and St. Paul Basilica in Rome when Gregory the Great was Pope in 599.716 Wood was used for constructing ships, buildings, crafts

713 After http://www9.georgetown.edu/faculty/jod/cassiodorus.html. 714 DE CARO 1985. 715 Large areas of the Sila mountains became ager publicus. The Romans caused severe damage to the hydrologic system of the region with their wild deforestation: PLACANICA 1999, p.56. 716 For the anecdote and bibliographical references regarding the transportation of the wooden beams from Bruttii to Rome, see FIACCADORI, p.734. 197

and for industrial needs. It was also treated to obtain pitch that was widely used for waterproofing ships and food containers, for aging wine, making cosmetics and drugs. Evidence from the first century AD shows that Brutti produced containers used specifically for the exportation of pitch.717 Perhaps pitch and timber extraction lasted until the last centuries of the Empire, but so far there is no evidence to confirm that these productions were strong even in late antiquity.718 These successful foresting activities were made possible in large part by navigable rivers that eventually dried up due to devastating deforestation.719 Mining might have also been another significant source of wealth in fourth century. Granite was quarried at Nicotera (south of Vibo Valentia) on the Tyrrhenian coast, located near the sea and convenient for transport.720 There were also important copper mines at Tempsa (modern Nocera Terinese), though these had been exhausted by the late first century B.C. (Strabo 6.1.5). F. Lenormant recorded traces of ancient silver mining at Verzino near Croton, and at Longobucco near Sibaris, but these have not been adequately documented. It seems probable, however, that the toponym Argentanum (modern San Marco Argentano) in the Valley of Crati near Cosenza refers to a silver-working center.721 Copper and chalcopyrite were also found in Bruttii. Mines have been discovered in the mountainous regions of Sila, Serre, and Aspromonte.722 Archaeologists found proof of metallurgic activity in Bruttii when they discovered a manufacturing establishment of iron dated between the fourth and fifth century AD in Santa Maria del Mare.723

A typical form of land organization in Southern Italy: the massa fundorum

It is the agricultural system based on the massa that, according to D. Vera, contributed to the rise of the economy in Southern Italy and produced new wealth for those who owned numerous estates in one particular region.724 The massa is a conglomeration of estates (fundi) of various size and type, not necessarily contiguous, which constituted a corpus within one civic territory and under the same municipal land taxation.725

717 DE CARO S., “Anfore per pece del Bruzio,” in Klearchos 27, 105-108, 1985, pp. 21-32. 718 NOYÉ G., “I centri del Bruzio dal IV al VI secolo”, in Italia Meridionale in Italia Tardoantica, Atti del trentottesimo convegno di studi sulla Magna Grecia, Taranto 2-6 Ottobre 1998, ibidem, 432; idem Popolamento, economia e società nella Calabria Bizantina (IV-XI secolo, in Storia della Calabria, III, edited by PLACANICA A., forthcoming; NOYÉ, G., “Les villes des provinces d’Apulie-Calabre, et de Bruttium-Lucanie du IV au VI siècle,” in BROGIOLO, G.P., (ed.), Early medieval towns, pp. 97-120. 719 On the effects of the Roman deforestation in Bruttii see PLACANICA 1999, p.57. 720 NOYÉ, G., 1998, see above note 170; PIPINO G., Oro, miniere, storia. Miscellanea di giacimentologia e storia mineraria italiana, Museo Storico dell’Oro Italiano, Ovada 2003, pp.55-61; LIPINSKI, A., Oro, argento, gemme e smalti. Tecnologia delle arti dalle origini alla fine del Medioevo (3000a.C.-1500 d.C.), Firenze, Olschki, 1975, pp.106-108 and 110-113. 721 LENORMANT, F., La Magna Grecia. Paesaggio e Storia, 2005. 722 NOYÉ G., “I centri del Bruzio dal IV al VI secolo”, in Italia Meridionale in Italia Tardoantica, Atti del trentottesimo convegno di studi sulla Magna Grecia, Taranto 2-6 Ottobre 1998, ibidem, 432; idem Popolamento, economia e società nella Calabria Bizantina (IV-XI secolo, in Storia della Calabria, III), edited by PLACANICA A., forthcoming. 723 ARSLAN 1998, p. 407 note 91; see NOYÉ 1998. 724 VERA 2000. 725 The fundus is not longer the production reference but only the fiscal datum for land taxation (cadastre):VERA 2001. 198

It was the prevalent form of land organization in whole Southern Italy and Sicily, throughout late antiquity, which improved the administration and management of rural estates and their revenues. 726 The massa would became a for social and religious geography of the territories. The massa favored landowners because it provided a guaranteed annual income. According to the Liber Pontificalis, owners of a massa earned between 100 and 280 solidi per year during the fourth century AD.727 The massa included a variety of cultivations according to the characteristics of the soil, and thus was able to provide a more stable revenue.728 For instance, if a wealthy landlord owned a massa in a territory that included plains, hills and mountains, he could divide productivity so that grain was planted on the plains, vineyard, olive and fruit trees were planted on the hills or fertile coastal areas, and wood was extracted on the mountains. This diversified approach to agriculture increased the potential for achieving substantial good returns from the land. The massa also allowed absentee landlords to collect rent revenue from occupying tenants, or coloni. Plenty of epigraphic and written sources mention that coloni were widely employed on aristocratic estates of Southern Italy throughout late antiquity. 729 Landowners could rent blocks of a massa to tenants rather than renting each fundus individually. The tribulations of early imperial landowners at managing their fundi and renovating their rental contracts with their numerous tenants attested by the Younger Pliny, was overcome by the new land organization of the massa.730 The massa appears in sources for the first time in the Vita Sylvestri who was pope from 314 to 335 AD. This text is part of the Liber Pontificalis,731 and although dated between 524 and 530 it records Constantine’s donations of privately owned estates to the Church during the early fourth century AD. Therefore, Vera believes that the massa developed earlier during third century.732 From texts such as the Liber Pontificalis733 and the Registrum of Graegorius Magnus, there is evidence of 60 massae distributed in central-southern Italy and Sicily.734 Literary sources do not mention massae in Northern Italy, but this does not mean that they did not exist there. Vera believes that the massa was well diffused in central and southern Italy, involving properties of the upper classes of society such as wealthy provincial landlords, Roman senatorial aristocrats, and high civic and military officials.

726 Sicilian evidence suggests that massae were geographically compact agglomerations of smaller fundi, often of substantial extent. See, ADAMESTEANU D., “Due problemi topografici del retrottera gelese,” Rendiconti Accademia Nazionale dei Lincei, Cl.Sc.Mor., St.e Filol., 8th series, 10 (1955), especially pp.205 ff. 727 Ibidem, p.621. 728 VERA 2001, BANAJI. 729 VERA 2000, p.621. On the rise of coloni, see C. Grey, “Contextualizing Colonatus: The Origo of the Late Roman Empire,” in Journal of Roman Studies 97 (2007), pp.155-175 with ample reference to earlier literature. 730 Pliny, Epistulae, IX, 37. 731 The Liber Pontificalis is a history of the popes beginning with St. Peter and continued down to the fifteenth century, in the form of biographies. See VERA D. , Osservazioni economiche sulla Vita Sylvestri del Liber Pontificalis. 732 VERA 2001, p.621. 733 Le Liber pontificalis, I-III, Duchesne L. (ed.), Paris 1955-57. 734 25 in Latium, 20 in Sicily, 7 in Campania, 3 in Calabria, 2 in Tuscany; 1 in Marche 1; 1 in Umbria and 1 in Puglia. See VERA 2000, p.615. 199

In Lucania et Bruttii there were three documented massae,735 but Cassiodorus writes that many massae existed just within the civic territory of Consilinum located in the Lucanian Val di Diano, near modern Padula.736 This suggests that very likely the massa was a well established form of land organization in Bruttii already in fourth century AD.737 The new agrarian system based on the massa probably contributed significantly to the increase of wealth in Bruttii during the fourth century AD. The estates of the rich landowning families of Locri were likely distributed between the fertile hills, the seaside plains and the mountains. Each area was exploited for the cultivation of various products according to the characteristic of the soil and terrain. Smaller crops, sheep-cattle rearing and extraction of timber occurred in mountainous areas. Apparently, the geographical diversity of the territory typical of the entire region of Bruttii was in a way heightened. The surplus of agricultural goods was placed on the market and commercialized effectively through ports and roads along the Tyrrhenian and Ionian coast.

The Presence of Senatorial Families in Fourth Century Bruttii

Plentiful natural resources and the potential for increasing wealth through trade with important Mediterranean markets and with Rome must have attracted senatorial aristocrats to invest their fortune in the estates of Southern Italy, including Bruttii. Moreover, this part of the Empire enjoyed relative peace during the third century while Central Italy was undergoing a major economic crisis (between 240 and 270 AD). As A. Carandini argues, already in the Antonine period the economic driver of the Empire was located at the center of the Mediterranean.738 Chastagnol points out that from the sources it appears that the strongest economic interests for the senatorial aristocracy were in Africa, Sicilia, and central-southern Italy. There lay the concentration of aristocratic wealth that supported their political and administrative interests.739 The economic rise of Bruttii was both reflected in and determined by the presence of powerful senatorial families in the region from the early Republican times, as attested by both literary and epigraphical texts.740 In the third century AD the Caerellii family, for instance had possessions in Gerace. The Cornelii and Uagellii families owned properties near Locri, and the Duc(eni)in Ardore.741 The influential and powerful gens Aurelia owned vast estates in Bruttii.742 One illustrious exponent of such families was Q. Aurelius Symmachus (fourth century) and later Cassiodorus (sixth century). Symmachus was corrector (governor) of Lucania et Bruttii in 365 AD while his father, L. Aurelius Avianius Symmachus, was prefect of Rome.743 He held the

735 Massa Nicoterana, Tropeana and Silana, see FIACCADORI 1994, p. 732. 736 VERA 1999b. 737 Ibidem. 738 CARANDINI 1986, pp. 7-9. 739 CHASTAGNOL 1987, pp.67-82 and passim. See also VERA D. 1988, p.131. 740 The presence of senatorial aristocrats during the Republican age is discussed in SANGINETO, “Trasformazioni,” p.206 ff; CAMODECA, G., “Ascesa al senato e rapporti con in territori d’origine, Italia. Regio I (Campania), II (Apulia et Calabria), III (Lucania et Bruttii),” in Epigrafia e ordine senatorio (Atti dell’AIEGL), Roma 1982, II, pp. 101-163. 741COSTABILE 1976, pp.81-85; FIACCADORI, G., “Calabria Tardoantica,” in Storia della Calabria Antica. Età Italica e Romana, II, ed. by S. Settis, Reggio Calabria 1994, p.738. 742CRACCO RUGGINI L., 1964; NOYÉ, G., 1998, p. 445. 743 MATTHEWS, J., Western Aristocracies, 1990, p.14. 200

governorship of Lucania et Bruttii probably during the same time when the Villa at Palazzi di Casignana was being renovated in grandiose style. It may be possible that he was corrector in the whole period until he was appointed to its next office, that of proconsul of Africa in 373 AD.744The seat of the corrector of Lucania et Bruttii was Rhegium together with Salernum, which was part of Lucania. Both cities were at the opposite end of an important and ancient maritime route that connected Bruttii with Campania.745 The important position of corrector held by Symmachus in 365, and by the governors who came before and after him, must have created a network of influential aristocrats not only in the city of Rhegium but also and in the nearby major centers of Locri, and Gioiosa which distant from Rhegium circa 100 km. Governors conciliated powerful local families while holding office, and they developed their own network of connections that extended across several provinces. The distribution of their estates all over the Empire facilitated this process. From the letters of Symmachus, we learn that he owned estates in many areas, including three suburban villas in Latium and 13 in Southern Italy. He visited the ones near Rome more regularly to assure upkeep of his property.746 He also owned income-producing estates in Samnium, Apulia, Sicilia, and Mauretania and Lucania.747 It is even likely that he had properties in Bruttii while he was corrector at Rhegium in 365 AD. Cassiodorus, who was also corrector of Lucania et Bruttii from 515 to 523 AD belonged to the same patrician family of the Aurellii as Symmachus. Cassiodorus is known to have owned estates in Bruttii and in his native Ionian town of Scolacium. Cassiodorus’ correspondent Venantius, also a corrector of Lucania et Bruttii, certainly owned properties in Bruttii. 748 The presence of senatorial aristocrats over the centuries and the existence of luxurious residences in Bruttiii suggest that the region offered opportunities to senatorial families to increase their wealth. Politics and entrepreneurial land ownership went hand in hand especially during late Roman times. The position of corrector could be occupied by senators and non- senators alike. Under Constantine, correctores of Lucania et Bruttii with aristocratic background were three times more numerous than the non-noble ones.749 Among them there were several relatives of the prestigious patrician family of the Aradii. The corrector Brittius Praesens, for instance, was related to the Aradii through the Valerii Maximi, and the Brittii were an old patrician family that dated back to first century BC. Another aristocratic corrector of Lucania et Bruttii was L. Turcius Apronianus (AD 323) whose two sons also became correctores of other Italian provinces (the former of Tuscia and Umbria, the latter of Picenum and Flaminia).750

744 The political career of Symmachus is typical of the Roman senatorial class. He was first quaestor, then praetor, before becoming corrector of Lucania et Bruttii in 365 AD. Afterwards he was proconsul of Africa in 373, praefectus urbis in 384-385 and consul in 391. See MATTHEWS 1975, RODA 1981, VERA 1986, CECCONI 1994. More recently, C. Sogno, Q. Aurelius Symmachus: A Political Biography, 2006. From LANÇON 2000, p.62: “Young people from senatorial circles could aspire to provincial governorship after which they normally travelled across the Empire to hold other offices. The most prestigious governorship was that of procunsul who could aspire to the top offices of vicarius of a diocese, the urban and praetorian prefectures. In Rome the high offices were that of the palace quaestor, master of the offices or count of finances.” 745 It is possible that the governor of Lucania et Bruttii travelled from one city to the other during the most favorable seasons for navigation and owned a residence in both cities. See GRELLE 1998, p. 134. 746 Symmachus, Epistulae 1.5;1.10;1.12. 747 VERA 1986, pp.231-275. 748 Cassiodorus, Variae III.8. 749 ARNHEIM 1972, pp.54-55. 750 Ibidem. 201

A. Giardina points out that the Diocletianic reform of the provinces opened even more than before new political career opportunities for the senatorial aristocrats in those provinces where they owned estates.751 The holding of the governorship could provide opportunities for establishing powerful patronage. Giardina mentions the case of Anicius Paulinus who was the first proconsul of Campania and a very wealthy landowner in that province. In one inscription it is claimed that through his political actions he contributed to move the image of Campania in prominent position within the hierarchy of the Italian provinces.752 The influential magistrates that emerged after the Diocletianic reforms became increasingly rich through investment in farming land. Their suburban villas were luxurious and grandiose.753 A governor of a province was seen by the landowners of that province as their defender against tax collectors of the state. He represented for the possessores a trusted officer versus treacherous officiales (officers) and tabularii (archive keepers) of the cities. Generally, the administrators of the state produced consensus among the upper classes who felt protected in their values and economic interests. 754 According to G.A. Cecconi, we should assume a correlation between the importance of a province and the political-social stature of the magistrate appointed to its governorship.755 From this perspective, the appointment in 365 AD of a prestigious senatorial aristocrat such as Symmachus to the governorship of Lucania et Bruttii is very significant. This points to the economic and social vitality of the region, and the availability of important resources in Bruttii during the fourth century AD. It is unlikely that a person of high political and social status such as Symmachus was appointed to rule a depressed and lethargic province of the Empire in which he could not have real opportunities to increase his wealth and political power. Symmachus’ presence in Bruttii suggests indeed the opposite. Very likely, this region was economically appealing even to the most illustrious members of the Roman senatorial class. Symmachus and his fellow governors likely had an entourage of aristocrats from Bruttii, as well as from other provinces and from Rome itself.756 As L. Lavan argues, the provincial aristocracy included several classes of people:

The provincial aristocracy can be roughly distinguished from the senatorial aristocracy as being part of the provincial land-owning elite without effective personal presence at court. This might, for example, include people with official rank of a senator, but who had never visited the capital. It further included the curiales, some notables emancipated from the curia, lawyers, some officiales, soldiers and the richer clergy: all who could be considered honestiores. This class typically owned land and maintained substantial estates, with a comfortable if not enormous domus in the city. In smaller

751 DURLIAT 1990, pp.14-29; 65-94; WHITTAKER-GARNSEY 1998; McCORMICK 2002, pp.64-119. CORBIER, M., “Coinage and Taxation: The State’s point of view, AD 193-337,” in Cambridge Ancient History, vol.12. The Crisis of Empire AD193-337, 2nd ed., in BOWMAN-GARNSEY-CAMERON (eds.), 2005, pp. 327-392, especially 360-386.. 752 For the inscription see AE 1972, 77; GIARDINA 1986b, p.2, note 183. 753 On the late Roman aristocracy and their life in the country see RODA 1985; CAMBI 1996; COLOMBI 1996; SFAMENI 1996; ARCE 1997. 754 GIARDINA 1986a, p. 27, notes 184-185. 755 CECCONI, G.A., Governo imperiale e èlites dirigenti nell’Italia tardo antica: problemi di storia politico- amministrativa, Como 1994. 756 A detailed analysis of the social hierarchy and officials’ duties in the Late Antique towns is found in CECCONI, G.A., “Honorati, possessores, curiales: competenze istituzionali e gerarchiche di rango nella città tardoantica,” in Le trasformazioni delle élites in età tardoantica, RITA LIZZI TESTA (ed.), Roma 2006, pp.41-64. 202

cities their number might stretch to include richer craftsmen who possessed the means and aspiration for social advancement.”757

Rhegium and the presence of the Roman elite on the Ionian centers

The largest number of late imperial luxurious villas, including the Villa at Palazzi di Casignana, are located at the main urban centers of the Ionian coast near Rhegium rather than in the northern part of Bruttii.758 Rhegium held the seat of the governor and was a major port on the Mediterranean. Senatorial families likely preferred to establish and maintain their residence near this important political center and harbor city. Moreover, Rhegium was easily connected with the rest of the Italian Peninsula through the Via Popilia (132 BC) along the Tyrrhenian coast linking Rhegium to Capuam.759 The relevance of Rhegium in fourth century is also confirmed by a number of inscriptions that attest to a consistent relationship between the local curia and the provincial government. From epigraphical and textual data, Rhegium seems to have had a more important role than Salernum, the other seat of the corrector. Imperial laws were published in Rhegium and important legislative archives were kept in this city. A group of ten laws addressed to the correctores of Lucania et Bruttii, dated between 313 and 365, were inserted later in the Codex Theodosianus, attesting to the important legislative role of the city.760 Other inscriptions record governmental measures of the correctores in regards to building temples and thermae in the city.761 The important jurisdictional role of Rhegium indicates that a large number of officiales (government officials) lived in the city or in the suburban villas nearby. They were most likely recruited from the local noble families rather than from the military. A funerary inscription, found in the nearby town of , refers to a young man named Centenarius, son of the bishop Leucosius, who is recorded as officialis.762 Moreover, the presence of a Jewish settlement in the city attested by an inscription and the existence of a rural synagogue discovered in the near town of on the Ionian coast,763 confirms that Rhegium constituted an important commercial and political pole also in relation with the eastern part of the Empire in fourth century.764 Olympiodorus, at the beginning of fifth century, states that Rhegium is the

757 LAVAN 2001; LAVAN-BOWDEN-MACHADO 2004. 758 SANGINETO 1994, pp.559-593 with tables I-V, and fig.17. 759 GIVIGLIANO 1994, pp.287 ff. Idem 1981, 1986, 1989. Part of the route of Via Popilia was found engraved on a stone called Lapis Pollae (The Polla Tablet). On this stone the road builder praised himself for having favored agriculture through the construction of the road, rather than cattle rearing. This is further evidence of the impact that the Via Popilia had on the development of the agricultural economy of the region. See also C. ADAMS- R. LAURENCE, Travel and Geography in the Roman Empire, (London ; New York : Routledge 2001), p.49. 760 GRELLE 1998, p. 132-133 with notes. 761 The inscriptions of Rhegium have been carefully studied by BONOCORE M., Regium Iulium, in Supplementa Italica 5, Roma 1989, pp. 29 ff. 762 1998. The inscription of Centenarius is ICI 5.8. 763 COSTAMAGNA L., “La sinagoga di Bova Marina,” in MEFRM, 103, 1991, pp. 611 ff. 764 Bruttii was strongly linked with the East. The prolonged use of the favored the commerce with the Eastern Empire. The relationship with the Eastern part of the Empire had an impact also on the Christianization of Bruttii. See PLACANICA 1999, p.64; OTRANTO 1998. 203

of Brettia (Bruttii), whereas four centuries earlier, Strabo had indicated Cosentia, located in the north of the region, as the main capital city.765 Unfortunately the Roman phase of Rhegium left behind little archaeological evidence. Several earthquakes wreaked the city over the centuries, especially the one in 1908, which destroyed ancient and modern structures. Furthermore, modern buildings have obliterated ancient ruins.766 It is in the rural areas of the Ionian coast near Locri and Gioiosa where more numerous Roman edifices have been discovered, as we have seen in Chapter 6.

765 GRELLE 1998. p. 134 note 56. The fragment by Olympiodorus is contained in FHG (Muller) 4, p.60, n.15=PHOTIUS 1 p.7, Henry. 766 For an account of the Roman city see PAOLETTI 1994, pp. 500-508. 204

CHAPTER 9: Room I inside the Villa at Palazzi di Casignana.

9.1. Room I at the centre of the villa’s eastern area.767

Room I is dated to the beginning of the fourth century AD and was built during the latest renovation of the Villa at Palazzi di Casignana (fig.9.1).768 It is by far the largest room so far uncovered inside the building (11.50 m. x 13.60 m.), and features a cruciform plan with one apse (ca. 6 m. wide) placed on the western side at the head of the cross. The apse and floor mosaics were severely damaged when road SS106 was constructed making it difficult to assert what was the function of Room I during the fourth century AD. The large size of Room I, positioned at the centre of Area D, its location on the villa’s main axis, as well as the sumptuous decoration composed of floor mosaics, wall marble revetments and frescoes, all reflect the importance and role played by this room as the main hall of the villa in the fourth century AD. The entrance to Room I is through a large opening at the

767 After Guida, p.80. 768 Room I was built during the R8 and R9 phases of the villa. See Guida, p.79. 205

centre of Portico P.769 Even the light that came from the portico must have contributed to create an effect of grandiosity.770 According to R. Krautheimer, rooms on the central axis and at the main entrance of imperial palaces (as Room I at Palazzi di Casignana), served functions that were closely related. They could be used as dining rooms or throne/audience rooms. They became an increasingly frequent component in the layout of the late Roman and early Christian palaces and palatial villas.771 Outstanding examples of centrally planned rooms include the Rotunda of Galerius in Thessaloniki located 125 m. northeast of the Arch of Galerius, which is now known as the Greek Orthodox Church of Agios Georgios, or simply The Rotunda (fig. 9.2). This cylindrical structure was built in 306 by the tetrarch Galerius and it was perhaps his mausoleum. In its original design, the dome of the Rotunda had an oculus like the Pantheon in Rome.772 Another circular room was built inside the smaller palace north of the hippodrome in Constantinople.773 A huge domed rotunda was erected in Diocletian’s palace at Split,774 and the decagonal one at Medica in Rome.775 An audience/throne hall existed in the Great Palace in Constantinople; it was domed and thus presumably centrally planned.776

9.2. Thessaloniki: The Rotunda of Galerius (4th c. AD).777

769 See Guida p. 81. This door was walled with rocks and coarse mortar perhaps after the villa was abandoned. The presence of an entrance to Room through Portico P is discordant with de Nittis’ reconstruction of the eastern façade, which shows a walled portico during the fourth century AD phase. See Guida, p.112. 770 The important role of lighting inside late Roman houses is discussed in ELLIS 2007. 771 See KRAUTHEIMER 1986, pp. 80-81. 772 DYGGVE, E., Recherches sur le palais imperial de Thessalonique, 1945; VITTI 1993. 773 Great Palaces in JOBST-VETTERS 1992. 774The huge domed rotunda in Diocletian’s palace at Split, according to Krautheimer, may be a salutatorium and audience hall. It is preceded by a colonnaded peristyle and raised on a podium. Its front is surmounted by the arched lintel of a fastigium, - a glorification façade. Perhaps, inside the room, the emperor received the high court charges and later he appeared to his subjects gathered in the tribunalium, an open-air area in front of the rotunda. 775 BOETHIUS, A., WARD –PERKINS, J.B., Etruscan and Roman Architecture, 1970, pp. 508 ff. 776 KRAUTHEIMER 1986, pp.80-81. 777 After http://en.wikipedia.org/wiki/Tomb_of_Galerius. 206

9.3. Room I: mosaic of the central area of the room.778

778 After Guida, p.84. 207

9.3b. Room I: Mosaic of the northern arm of the room.779

The mosaic that paved Room I is preserved only in the central and northern parts of the room. The central area (fig. 9.3) features a polychrome motif of small tesserae divided in different sections. The northern arm of the cross plan (fig. 9.3b), on the other hand, is decorated with a black and white composition of hexagons inserted with six-pointed stars.780 Even the walls of Room I were lavishly adorned. They were covered with marble revetments (crustae) that likely were very high, reaching the ceiling.781 Portions of the marble slabs and of the bronze cramps still stand in place today.782 The upper parts of the walls were perhaps framed with coloured bands since many fragments of painted plaster were found in the rubble of the room. Room I was not heated in the fourth century AD, although tubuli are currently visible on the extant portions of the apse’s walls. They belonged to the heating system and praefurnia of the

779 After Guida, p.85. 780 Décor, v.2, p. 41. An analysis of these mosaics is on Chapter 5 of this study. 781 The marble blocks were cut into thin slabs with saw-mills. Various kinds of sand were used in the operation, according to the hardness of the stone; emery (naxia, Pliny, NH, 36.6 s9) being used for the hardest. This technique was probably Oriental in its origin. The brick walls of the Mausoleum at , built around 355 B.C., were covered with slabs of Proconnesian marble (Pliny, NH, 36.6), and this is the most ancient example on record. In the time of Pliny (NH, 35.1) slabs of a uniform color were sometimes inlaid with variously colored materials to represent animals and other objects. (Domus p. 431; Pictura, 15). 782 Pliny NH, 36, 47-50 writes that marble crustae may have been invented in around 350 BC but reached Rome only three centuries later when the technique was used to decorate private residences of the very wealthy. Pliny is in general critical of the use of marble in private residences. The limited supply of marble in Rome, until the end of the Republic, promoted the popularity of veneers among the wealthy Romans. Veneers resembled solid marble but were less costly. J.W HUMPHREY- J.P. OLESON– A.N. SHERWOOD, Greek and Roman technology: A Sourcebook, 1998, p.250. 208 previous third century AD rooms that lie underneath Room I. It is possible that Room I had a conventional tiled roof supported by timber rafters rather than a concrete barrel vault. The walls of Room I were in fact dissimilar in construction technique because they were partly taken from the earlier third century structures. Thus, they were not perfectly orthogonal and not solid enough to support a heavy dome roof.783 The following sections describe the development and function of apsidal halls in Roman architecture, and discuss three possible functions of Room I during the fourth century AD. By outlining the most common uses for the larger apsidal halls such as Room I, we can better understand the significance that apsidal rooms possessed as spaces during this period.

783 See Guida, p. 85. 209

Apsidal Halls and Their Development

The term ‘apse’ refers to a semicircular recess where the floor is on the same level as the rest of the room and where the roof is the section of a dome. The problem of the origin and meaning of the apse has been the subject of several studies including the detailed work of B. Tamm who underlined its sacred connotation.784 According to Tamm, the earliest known apses were associated with grottoes and fountains, but by the early Empire they were incorporated into public buildings. J.B. Ward- Perkins contends that the apse made its formal appearance in the first century AD inside the Palace of Domitian, or Domus Augustana, on the Palatine (fig. 9.4) as the setting for the semi- divine majesty of the .785 9.4. The Domus Augustana (92 AD).786

During the fourth century AD, apses became a common feature in both public and private architecture. A careful study on the late Roman domestic architecture by I. Baldini Lippolis, suggests that when an apsidal room is connected to the peristyle, on the main axis of the building and in a prominent position, it likely served the function of a dining room and/or a reception room. If one house featured several reception rooms, which are more or less equal in importance, then the apsidal room could have functioned mainly as a reception room where people gathered for meetings rather than for dining.787

784 This scholar studied the apse and its meaning from the Augustan to the Flavian period, see TAMM, B., Auditorium and Palatium, Stockholm 1963, pp.17-216, especially p.57. The apsidal triclinium in the Domus Augustana or Palatium (built by the architect Rabirius for the emperor Domitian and inaugurated in 92 AD), according to Tamm provided the opportunity for the emperor to present himself as a god. Apsidal rooms and their symbolic meaning have also been treated by DUVAL 1984, pp. 129-152 and 447-470; GUIDOBALDI 1986, pp.165-237; SCAGLIARINI CORLAITA 1995, pp.837-873; MORVILLEZ 1996, pp. 119-158; ELLIS , S. P., “Late antique dining: architecture, furnishings and behaviour,” in Domestic Space in the Roman World. Pompei and Beyond, in JRA, Suppl. Ser., 22, 1997, pp. 41-51, and 129-152; BALMELLE 2001, pp.155 ff. 785 WARD-PERKINS, J.B., Architettura Romana, Milano (Electa) 1974 , pp.66 ff. 786 Ibidem. 210

At Palazzi di Casignana, we have hypothesized that Room 7 (see plan of the villa in Chapter 4) located near the baths of the villa was a dining room, or triclinium, since it was elegantly decorated and placed in a prominent location looking out into the large peristyle of the villa (Courtyard 55). It may be that Room I, instead, was used for holding large gatherings rather than for dining. It should not be forgotten that rooms in Roman domestic architecture were often multifunctional. For this reason, fixing one particular label to a room may not be always correct.788 In exploring the possible functions of Room I during the fourth century AD, this study presents a provocative hypothesis that this space hosted private and maybe even public religious gatherings. Based strictly on its architectural characteristics the three functions of Room I may have been: 1) A dining room 2) An audience hall or 3) A private or semi-private Christian chapel.

787BALDINI LIPPOLIS 2001, p. 59. For a survey and bibliography of domestic architecture in late antiquity across the Empire, see SODINI 1995; 1997; ELLIS 1991; 1997; 2000;ELLIS 1995 (Britain); ROMIZZI 2006 (Italy); CHAVARRIA-ARCE-BROGIOLO 2006 (Spain). For a treatment on the architecture of houses, villas and palaces, see SCAGLIARINI CORLAITA 2003. 788 ROSSITER 1991; ELLIS 1991; SODINI 1995 and 1997. We know from the sources, for instance, that inside the cubicula (bedrooms) the Romans also dined or held important political meetings. 211

1) Room I as a dining room (triclinium).789

9.5. Reconstruction of a wooden stibadium based on a mosaic representation found inside the Villa of Falconer at Argos (6th c. AD).790

From the second century AD onward, the apse was used to receive a semicircular dining couch, a stibadium,791 which came to replace the rectangular couches (triklinai: three couches) of the Republican and early imperial period (fig.9.5).792 The stibadium was known also as the sigma-couch because its semicircular shape resembles the Latin capital letter C that corresponds to the Greek lunate sigma.793 As K. Dunbabin points out, the stibadium was originally used for outdoor banqueting while hunting or during religious celebrations.794 The semicircular couch was placed directly on the ground with a low table or a plate set in its curve as it is represented on the Small Hunt mosaic from Piazza Armerina (fig.9.6).

789 See DUVAL 1984, pp. 447-70; ROSSITER, J., ‘Convivium and Villa in Late Antiquity,’ in SLATER, W.J. (ed.), Dining in a classical context, Ann Arbor 1991. 790 After BALDINI LIPPOLIS 2001, p.80, fig.21. 791 On masonry stibadia see DUVAL 1984, pp. 459-460; MORVILLEZ 1996, pp.116-158; VOLPE et alii 2004 and 2005a and 2005b; DUNBABIN 2003, pp.169-174. On the relationship between stibadia and aristocratic domestic architecture in late antiquity see DUNBABIN 1991, eadem 1996; ROSSITER 1991, especially pp. 202-206; ELLIS 1991; idem 1997, in particular pp.41-42, 46-51; idem 2000, pp. 148-150. 792 ELLIS, S. P., ‘Classical Reception Rooms in Romano-British Houses,” in Britannia, 26, 1995, p. 119. See BEK, L., “Quaestiones Convivales: The Idea of the Triclinium and the Staging of the Convivial Ceremony from Rome to ,” Analecta Romana Instituti Danici, 12 (1983), pp. 71-105. See also MALMBERG 2003. 793 The origin of the Greek word stibadium is stibades, i.e. the foliage gathered together on the ground for lying on, or used to stuff mattresses. See DUNBABIN 2003, pp. 146 ff. 794 Outdoor banquets are represented on a silver plate from Cesena (northern Italy) and on the Hunting Plate from the Sevso Treasure. See DUNBABIN 2003, pp.141-150. 212

9.6. Piazza Armerina: The Small Hunt Mosaic (320-330 AD). A portable stibadium is laid on the ground during an outdoor meal.795

This custom migrated indoors by the later centuries of the Empire and became predominant especially in the fourth century AD, as attested in many wealthy houses all over the Roman world.796 Apses in reception rooms of the late Roman houses varied in size. Some could fit one couch and provide a space for service behind; in others the couch would be placed directly

795 After CARANDINI- RICCI- DE VOS 1982, pp.175-178, Room 30, pl.24. 796 DUNBABIN 2003, pp.141-150. 213

against the wall leaving no space behind.797 The couch could offer room for five or more diners, depending on the size of the apse in which it was placed.798 Sidonius Apollinaris describes a banquet offered in Arles by the emperor Majoranus (457-461 AD) and illustrates the arrangement around the table of the eight diners. The emperor sat at the extreme right (in margine dextro), whereas the most prestigious guest sat on the opposite end, at the extreme left (cornu sinistro), in front of the emperor. All other guests sat in order from left to right in accordance to their social level and importance. The least desirable seats were those behind the emperor’s shoulders because they provided less visibility of the prestigious host.799 At the Villa of Palazzi di Casignana, the apse of Room I is placed in the back western wall of the room in central position and is circa 6 m. wide. It may have held a stibadium for 6 diners. The two rectangular arms of the cruciform plan may also have been used for a dining setting for a triclinium of the early imperial type. Each wing measuring 5 x 3 m. could in fact receive three couches (triklinai) arranged in the shape of the Greek letter Pi.800 As noted above, the arms of the cruciform plan are marked off by mosaics with different geometric compositions. The central area of Room I would have offered a large space for serving food, gatherings or entertainment to which the diners would be spectators. This proposed organization of the space in Room I suggests a hierarchy in seating arrangements. The stibadium couch on the apse would have had greater prestige and likely accommodated the dominus of the villa, his wife and their most distinguished guests; whereas the lateral wings of the room (the arms of the cruciform plan), being remote from the centre of the room and providing a reduced view of diners and entertainment, would have held the dining setting of less important guests. In a dining setting such as the one proposed for Room I at Palazzi di Casignana, there was less contact and communication between most of the guests, depending on where they were seated, which correlated with their social status.801 As L. Bek shows, the apse in the third and fourth century AD functions as a backdrop, not for sculptures and fountains, but for people and events. The effect was to glorify and give high prestige to the master of the house. She writes:

“Instead of being an entertainment of the participants, the convivium has become a performance to be staged. The display of social interrelation has become an official show or state ceremony.”802

797 Ibidem, pp.169-174. 798 See dimensions in MORVILLEZ, E., “Sur le installations de lits de table en sigma dans l’architecture domestique du Haut et du Bas-Empire,” Pallas 44, 1996, pp.119-158, especially pp. 137-139, 158. 799 Sidonius, Carm., 1.12. 800 From DUNBABIN 2003, p. 216 note 6: “the term triclinium is used for both the arrangement of the three couches and for the room capable of containing such arrangement; the former is evidently the primary sense. In later Latin, it comes to mean simply the room in which one dined, regardless the nature of the couch arrangement. For the sense of an arrangement of three couches, see e.g. Varr. LL 9.4.9. LEACH 1997, 67-68, discusses briefly the literary sources for triclinium (dining room) and the other main Latin term for a dining room, cenatio, which she believes usually designates a large and grandiose room.” 801 On seating and social status, see ROLLER 2006. 802 BEK 1983, p. 91; SCOTT 2004, p. 52. 214

The banquet within an imposing apsidal space and lavishly decorated setting offered to the villa’s patrons a theatre for playing out their social roles and interacting with other high society acquaintances. Representations of stibadia and tables are found in funerary paintings of the fourth century AD (fig. 9.7),803 and also on manuscripts dated to the sixth century AD.804

9.7. Banquet scene around a stibadium from the “Tomb of the Banquet” in Constanza (mid-4th c. AD).805

803DUNBABIN 2003, pp. 176-202. There are several examples including the paintings inside the catacombs of Callisto (GRABAR 1980, 1, fig.105), and that of Ss. Pietro and Marcellino (GRABAR 1980, 1, fig.111). 804 See Codex sinopensis in GRABAR 1980, 2, fig. 227. 805After http://www.ascolisatrianofg.it/ascolisatrianofg/siti_archeologici_faragola.htm. 215

9.8. Detail of nave wall mosaic in the Church of St. Apollinare Nuovo, Ravenna, depicting the Last Supper. Glass and gold-leaf (6th c. AD).806

In the Church of St. Apollinare Nuovo in Ravenna a sixth century mosaic represents the Last Supper with a stibadium (fig. 9.8).807 Dining scenes usually refer to royal or religious context. The fourth-fifth century AD villa at Faragola in Puglia (Apulia) provides an outstanding example of masonry base for a stibadium that was decorated with luxury opus sectile inlaid decoration (figg.9.9 and 9.10). A comparable masonry stibadium was found in the Villa at El Ruedo in Baetica dated between the third and fourth century AD.808

806 After: http://www.sacred-destinations.com/italy/ravenna-sant-apollinare-nuovo-photos/slides/xti_6926p.htm 807 DUNBABIN 2003, p.201, fig.120. 808Ibidem, pp.169; VAQUERIZO-CARRILLO DIAZ-PINES 1995, pp.145-147. 216

9.9 and 9.10. Villa at Faragola in Puglia (mid 4th c. AD). At the top: the ruins of the cenatio with an extant masonry stibadium. At the bottom: the three-dimensional reconstruction of the cenatio, stibadium, opus sectile floor and its splendid geometric panels along the central axis of the room.809

809After: http://www.archeologia.unifg.it/ric/scavi/far.asp. These panels mimic the quasi-emblemata technique of inserting multiple panels into mosaic floors from the 2nd century AD onwards. The geometric panels of this kind permits the viewer to admire them from any angle. 217

In several houses and villas dated between the fourth and sixth century AD there are reception rooms equipped with three apses. These are called triconchs. The triple number of the apses inside the triconchs has been interpreted as a desire by the wealthy patrons to maintain the tradition of having three couches.810 The three apses could contain a total of three stibadia, while the central space was left open or used for entertainment.811 Triconchs also vary considerably in size. Several of them have been found in Tunisia at Thuburbo Maius,812 and all over the western provinces of the Empire. Monumental ones were built for instance at the Sicilian villas in Piazza Armerina and Patti Marina;813at the villa of Desenzano in Northern Italy (figure 9.11),814 and at several villas in Aquitania.815

9.11. Villa at Desenzano: triconch (mid-4th c. AD).816

At Palazzi di Casignana, Room I lies right behind Portico P of the eastern façade suggesting that this hall was not reserved exclusively to the intimate friends or acolytes of the villa’s dominus, but had a more public function, likely that of hosting large gatherings. For its dimensions, Room I might have contained approximately 100 seated people.817 In several late Roman villas, including Piazza Armerina (Sicily), Desenzano (northern Italy), Löffelbach

810 ELLIS, S., “Power, Architecture, and Décor: How the Late Roman Aristocrat Appeared to His Guests,” in Roman Art in the Private Sphere, 1991, p.119. 811 Literary sources of the second century AD refer that the ideal convivium should have no more than nine guests, the number of the . Yet, as K. Dunbabin shows, from the end of the first century onwards, in Pompeii developed a tendency toward big dining rooms capable of holding more than the traditional nine guests but still arranged in the Pi shape of the traditional early imperial triclinium arrangement. DUNBABIN 2003, pp. 39 ff. For a survey of triconchs in Late Antiquity see MORVILLEZ 1995. 812 BEN ABED-BEN KHADER, Thuburbo Maius: Les mosaïques dans le région ouest, (M. Alexander and M. Ennaifer eds.), Corpus des mosaïques de la Tunisie, 2, 3, Tunis 1987. 813 For Piazza Armerina see WILSON 1983a; for Patti Marina 1976-77, pp.575-579 and BACCI 2001. 814 SCAGLIARINI CORLAITA 1997b. 815 BALMELLE 2001, passim. 816 After GHISLANZONI, G., La villa romana di Desenzano, 1962. 817 This figure is based on the capacity of modern conference rooms that measured 100 square meters. 218

(Austria), and Thuburbus Maius (Tunisia), the main apsidal halls were placed at the end of a ceremonial way. At Piazza Armerina, for instance, before reaching the basilica, one needed to cross a monumental vestibule, a peristyle, then the double apsed gallery of the Great Hunt, in order to approach the audience hall. Similarly, at the Villa in Desenzano, the trichonch was reached after crossing a first octagonal vestibule, a large peristyle, and a double apsed atrium.818 Because of its location, Room I of the Villa at Palazzi di Casignana may have been a place for large gatherings rather than a more exclusive ceremonial reception room. This possibility is discussed in the following section.

818 SCAGLIARINI CORLAITA 1992, pp. 54, ff.; WILSON 1984. 219

2) Room I as an Audience Hall.

Room I of the Villa at Palazzi di Casignana may have functioned as an audience hall for the dominus, a local magnate, to whom people presented their petitions or concerns and asked for support. The apse could have provided a sumptuous background for the seat of the dominus creating an atmosphere that emulated the emperors’ monumental audience halls.819 Large apsidal halls were built inside imperial palaces such as the Palace of Domitian on the Palatine in Rome,820 and they were used as audience halls or throne rooms with the ultimate purpose of glorifying the emperor. The most remarkable example of an imperial audience hall is represented by the Basilica of Constantine or Aula Palatina at Trier (Germany). This was built by the emperor Constantine around 310 AD as a part of the palace complex and contains the largest single inner room surviving from antiquity (see fig. 9.12).821 The Aula Palatina has a length of 67 m., a width of 27 m. and a height of 33 m.822 There, Q. A. Symmachus presented the Crown Gold (aurum coronarium) to Valentinian I in 369 AD.823 The apse framed the emperor and created a magnificent setting for his person. It may also have possessed religious symbolism. P. Heater describes the emperor as sitting in the apse at the far northern end of the hall with the dignitaries of state lined up in front of him in order of seniority. The hall was likely richly decorated with marble statues and luxury fabrics.824 In emulation of the emperors, from the second century AD onward, and especially in the third and fourth century, very wealthy and notable Romans across the Empire reproduced palatial audience halls at smaller scale inside their residences. Apses became essential elements in aristocratic houses, usually given extra emphasis by their impressive dimensions and sophisticated decoration. They conferred a regal and grandiose aspect on the room lending glamour and prestige to the persons occupying them.825 Yet simple emulation was not the main cause of the spread and adoption of large apsidal rooms. More importantly, it was the social and political change that took place during the late Roman Empire that determined the wide diffusion of these rooms in the private sphere.

819 For instance, the aula regia in the Domus Augustana. See WARD-PERKINS 1974, pp. 66 ff. 820Ibidem. The Palace of Domitian (Domus Augustana) represents a milestone in the history of Roman architecture. For the first time a single complex included all the functions of the state’s public life, organizing them in a harmonious and functional manner. Within the palace, the separation between the public and private structures responded to the need for isolating and highlighting the person of the emperor. He was the semi-divine figure who appeared to the public only in specific monumental areas of the complex. The role of the apses was fundamental for this purpose. They bestowed authority and a sense of divinity to the emperor framing his ‘divinity’ in front of the mortals. See P. Zanker, “Domitian’s palace on the Palatine and the imperial image,” in A. K. Bowman et alii (ed.), Representations of the Empire. Rome and the Mediterranean World (Oxford 2002), pp.107-130. 821 ZAHN, E., Die Basilika in Trier. Rheinisches Landesmuseum, Trier 1991. 822 In 1856 it became the protestant Church of St. Savior, see ibidem. 823 HEATHER 2006, p.22: “Crown gold was a theoretically voluntary cash payment which the cities of the Empire handed over to emperors on their accession and on every fifth anniversary (quinquennalia) subsequently.” 824 Ibidem, fig.4. 825 GERVASONI-LAMOINE-TRÉMENT 2004. 220

9.12. The Aula Palatina at Trier, Germany (310 AD).826

826 After HEATHER 2006. 221

The rich and powerful upper classes had large and important clientelae.827 These influential relationships were often handed down from generation to generation. Wealthy landowners who exercised authority in a particular area or who had direct access to, or some connection with, the imperial court at Rome or Constantinople, were often approached by their clients for advice, to obtain favours and even arrange protection. As S. Ellis states:

“Late antiquity was a time when power became more nakedly associated with the individual. The breakdown of local government made it more effective to solve problems through a man of power with influence at court than through legitimate bureaucratic routes. The emperors since Diocletian had consciously adopted more ceremonial, making themselves seem remote, untouchable and impassive. Local magnates naturally followed the role model of the emperor.”828

During the late Roman centuries, the local governments of municipia increasingly diminished in power. As a consequence, the common people came to rely upon the protection of strong patrons rather than the law or the city council.829 Local leadership and patronage was frequent in the western provinces.830 Whenever average citizens needed a strong protector to solve problems they would turn directly to local personages rather than more distant centres of authority. The late Roman state encouraged the concentration of power in the hands of local grandees by appointing them to positions of authority over their homelands.831 The need for audience halls and grand dining rooms seems to fit well within this specific social reality. Built by the wealthiest and most powerful, these halls hosted copious banquets, aristocratic receptions and were capable of receiving large number of poorer clients from the local community.832 The imposing way in which powerful domini presented themselves to their clients and guests in their monumental dining-audience-halls imitated the way the emperors appeared to their official audiences as untouchable and unreachable figures. Ammianus Marcellinus illustrates the adventus of Constantine II at Rome during his ceremonial arrival in a manner that is explanatory of the same custom adopted by the wealthy Romans at the end of the Empire.833 In suburban villas, apsidal rooms may range in length between 20 and 30 meters (Room I at Palazzi di Casignana is 11 m. long).834 They are often designated as basilicae, on the basis of a passage from the Historia Augusta that mentions three basilicae that were 100 feet long (30 m.) inside the villa of Gordianus near Rome.835 Apsidal rooms are usually considered the most

827 ELLIS 1994; BROWN 1987, pp. 371-374; ROULAND 1979. 828 ELLIS 1988, p. 134. 829 For a discussion on this subject, see ELLIS 1988, p.41; CAMERON 1993, pp. 93-94. 830 VAN DAM 1985, especially Part I, pp. 14-16. 831 ELLIS 1981, p.41. 832 ELLIS 1995, p.190. 833 Ammianus Marcellinus, 28.4.12. ELLIS 1991, pp.129-130. This practice began with the introduction of eastern monarchic ceremonies in the third-century, which elevated emperors to a transcendent sphere. 834 CARANDINI et Alii, 1982; DUVAL 1987, idem 1992; BALMELLE 2001. 835 BALMELLE 2001, p. 171; Historia Augusta: Gordianus 20.32. p.54; LUSCHI 1982, pp.157-178. 222

prestigious areas of the residences, and are normally taken as evidence that the owner was a powerful and influential person. In Italy, an outstanding example of an enormous apsidal hall, or basilica, is found inside the Villa at Piazza Armerina (fig. 9.13),836 which can offer an analogy for the kind of spaces that the late Roman aristocracy and upper classes built in their residences for self-representation during the fourth century AD. It covers an area of 440 square meters and measures about 30 m. x 18 m., falling in the category of the classic basilica centenaria, being a hundred Roman feet long.837 Scholars have attributed to this enormous room the function of an audience-reception hall, or that of a triclinium.838

9.13. The Basilica inside the Villa at Piazza Armerina (mid-4th c. AD).839

836 CARANDINI et Alii, 1982, tav.III. 837 WILSON 1983, pp.25, 91. 838 According to Wilson this hall was probably designed for hearing petitions from clients or for receptions and banquets: WILSON 1983, p.25. On the contrary, Ellis believes that it served as a triclinium because “it does not share the characteristic street-side position of audience chambers in the ‘palaces’ of Asia Minor. Rather, it is axially aligned with the central peristyle of the villa”: ELLIS 1994, p. 122. 839 After CARANDINI et Alii 1982, pl.3. 223

Another surviving large apsidal hall of 300 square meters was built in Ravenna inside the so called ‘Palace of Theodoric,” which is not a villa but a town home built in the fifth century AD (fig. 9.14, room T).840

9.14. Ravenna: Palace of Theodoric and its basilica (Room T, 5th c. AD).841

840 BALDINI LIPPOLIS 2001, pp. 253 ff.; DUVAL 1960. 841 After ELLIS 1991, fig.6-2. 224

9.15. Rome: House of Junius Bassus (mid-4th c. AD). 9.16. Rome: House of Sette Sale (4th c AD).842

842After GUIDOBALDI 1986. 225

9.17. Villa at Montcaret: the cruciform hall (4th c. AD).843

Comparable examples of great apsidal halls are found in Rome inside the fourth century homes, including the House of Junius Bassus (fig.9.15) on the Esquiline Hill dated to 331 AD,844 and the House of Sette Sale (fig. 9.16).845 In France, the fourth century large basilica at Villa at Montcaret (330 square meters) features a cruciform plan with an apse at the two opposite heads of the cross (fig. 9.17).846 The Villa of Nerac in Aquitania also contains a large apsidal hall that measures 240 square meters.847 Several other examples may be cited in the same province, also dated to the late imperial times.848 In Spain, a luxury villa at Cordoba contains an enormous apsidal hall that is 45 m. long, with an extension of 700 square meters (fig. 9.18).849

843 After BALMELLE 2001, II, p. 170, fig.66. 844 ASHBY-LUGLI 1932, pp. 221-255. 845 GUIDOBALDI 1986, pp. 167-171; Idem, 2000 and 2004; BIANCHI-BRUNO-DE NUCCIO 2002. 846 BALMELLE 2001, p.171. 847 Ibidem. 848 ARCE 1997a. 849 HIDALGO PRIETO 1996, pp. 24-25, figg.4-5 and figg. 63-72. 226

9.18. Spain: Villa at Cordoba (4th c. AD).850

850 After HIDALGO PRIETO 1996, figg.4-5 and figg. 63-72.

227

9.19. Villa at Desenzano in Northern Italy (mid-4th c. AD). The apsidal Room 35 is part of Area B.851

Alongside these huge examples, which may be termed basilicae, there is a series of large apsidal rooms found in suburban villas that measure more or less 100 square meters. Room I of the Villa at Palazzi di Casignana belongs with these. In Italy, such apsidal halls are found, for instance, inside the late Roman villa at Desenzano on the shores of Lake Garda (fig.9.19),852 and at both the Sicilian villas at Patti Marina (fig. 9.20) and Tellaro (fig. 9.21).853 In Rome, smaller apsidal rooms become frequent in the houses of the elite after the Severan age (235 AD), although they are not represented on the Forma Urbis.854 Small-scale basilicae are also found in France at the Villa of Seviac,855 in Spain at the Villa of Aguilafuente856and at Carranque (fig. 9.22). 857

851 After SCAGLIARINI CORLAITA 1992, fig. 13. 852 SCAGLIARINI CORLAITA 1992, pp. 80 ff. 853 VOZA G. 1976-1977, L’Attività della Soprintendenza alle antichità della Sicilia orientale. Parte II, Kokalos 22- 23, pp. 551-585. Idem, 1980-81; 1982;194-85; 1999, 2003. 854 The Forma Urbis is the massive marble map of ancient Rome created under the emperor Septimius Severus between 203 and 211 AD. It was 18 m. wide and 13 m. high. It was carved into 150 marble slabs and mounted on an interior wall of the Temple of Peace. See GUIDOBALDI 1986; BALDINI LIPPOLIS 2001, p.59. 855 BALMELLE 2001, p. 109, fig.44. 856 FERNÁNDEZ-CASTRO 1982, fig.4. 857 FERNÁNDEZ-GALIANO 1994, pp.317-324, figg.1-2. 228

9.20. Sicily: Villa at Patti Marina (4th c. AD).858

858 After WILSON 1983, p.73. 229

9.21. Sicily: Villa at Tellaro (mid-4th c. AD).859

5.22. Spain: Villa at Carranque (late 4th c. AD).860

859 After WILSON 1983, p.73. 860 After BROGIOLO-CHAVARRIA ARNAU 2004, fig.12. 230

9.23. Portugal: Villa at São Cucufate (4th c. AD).861

861 After ALARCÃO et Alii 1990, p.123, pl. 70.

231

9.24. 1) The Palace of (4th c. AD). 2) The Governor’s Palace at (4th-5th c. AD). 3) The Palace of Dux at (early 6th c. AD).862

There are other numerous examples of apsidal halls in domestic architecture that have been identified as audience halls. They include those at the Palace at Ephesus (Turkey) dated to the early fourth century (fig. 9.24,1),863 the Governor’s Palace at Aphrodisias (Turkey) dated to the late Roman period (fig. 9.24,2),864 and the Palace of the Dux in Apollonia in Cyrenaica (modern Susa, Libya) dated to the early sixth century (fig. 9.24,3).865 Semicircular or rectangular niches are frequently cut out on the back wall of the apses, perhaps to place small statues or other monumental elements.866 It is possible that even the apse’s wall of Room I at Palazzi di Casignana, not longer extant, had niches where statues were

862 After ELLIS 1991, fig.6-1. 863 ELLIS 1988, p. 569; SCOTT 2004, p.52. 864 GOODCHILD 1976, pp.245-265; ELLIS 1985, pp.15-25. 865 ELLIS, S., “The Palace of the Dux at Apollonia and Related Houses,” in BARKER, G., LOYD, J., and REYNOLDS, J., (ed.), Cyrenaica in Antiquity, Oxford 1985, pp.15-25. 866 See houses at Amphissa and Athens in BALDINI LIPPOLIS 2001, p. 59. 232 placed. The discovery of glass tesserae in the room’s rubble suggests that the apse was decorated with a glittering mosaic. Such sumptuousness was typical of late Roman villas in general, and was widespread across the Empire - especially in the capital cities of Rome, Trier, Arles, Milan, Ravenna, Antioch, Thessaloniki, and Constantinople.

233

3) Room I as a chapel.

So far we have seen that Room I inside the Villa at Palazzi di Casignana could have been used as the most formal dining room of the residence and/or as the private audience hall where the powerful dominus met regularly with people living in the area. A third possibility is that Room I may have also hosted religious gatherings. The interlocking of the private and public, pagan and Christian, imperial and the religious sphere could actually be typical for late Roman imperial palaces and villas. Studies by R. Krautheimer show that all late Roman triclinia and basilicae likely served both religious and secular functions as implied in the position of the God-Emperor. In domed rooms, in particular, the idea of the Dome of Heaven was ever present. The Golden Triclinium inside the Great Palace in Constantinople likely served both religious and secular function: it was an audience hall for the Emperor’s Majesty and also a palace chapel.867 No traces remain of it, but ancient texts describe it as having a circular plan, enveloped by eight vaulted niches, which opened into adjoining rooms. Krautheimer points out that churches inside the imperial palaces in the fourth century AD were probably more numerous than we imagine. Although none have been identified with certainty, there are numerous candidates.868 If the dominus of the Villa at Palazzi di Casignana were Christian, Room I could have served the same function as the church inside imperial palaces where the Emperor, God’s counterpart on earth, would attend religious services.869

867 KRAUTHEIMER 1986, p.81 note 29. 868 Ibidem 869 Ibidem 234

Excavations at Palazzi di Casignana have not produced secure evidence of cultic use. Therefore the possibility that this room may have functioned as a private or semi-private chapel in the fourth century AD remains for the moment in the realm of speculation. Although it would be safer not to move beyond the concrete material data that has so far emerged from the excavation, it is worth taking a less conservative approach, and looking closely at all the

9. 25. Eastern sector of Villa at Palazzi di Casignana.870

architectural changes that occurred in the eastern sector of the villa (fig. 9.25), which may be related to a possible Christian function of Room I. For taking all together, these changes may invite us to consider precisely this possibility.

a) The presence of a third century AD octagonal room beneath the cruciform Room built in the fourth century AD. Both rooms may have served a religious function since the octagon and the cross both convey a symbolic Christian meaning. b) The cruciform plan of Room I and its east-west orientation are two characteristics typical of early Christian churches.871 c) The transformation of Room E (see fig. 9.25) into a water reservoir in the fourth century AD may have been dictated by liturgical needs for the Christian practice of baptism.

In addition to these architectural modifications two further considerations frame our interpreting Room I as possessing a religious function:

d) The presence of a large number of burials around this part of the villa, while it was still inhabited, which may be connected to Room I having served as a Christian chapel

870 After Guida, p.80. 871 MACKIE 2003. 235

e) The existence of a developed Christian community in the territory of Locri (and Bruttii) between the third and fourth century AD when Villa at Palazzi di Casignana reached its splendour.872 The fact that Locri became a Diocesis in 330 AD encourages us to consider that the Villa at Palazzi di Casignana may have contained a space for Christian gatherings and cult. The architectural changes to Room I and the presence of burials on the villa’s site will be discussed in the following sections, whereas the diffusion of Christianity in Locri and Bruttii will form the subject of Chapter 10.

872 The diffusion of Christianity in Bruttii is discussed in Chapter 10 of this study. 236

9.26. The octagonal Room V underneath Room I.873

873 After Guida, p.76. 237

a) The octagonal Room V. Room I and the southern rooms of the eastern sector of the villa (M, N, O, P1, Q, R, S, Z) were all built on top an earlier heated set of rooms dated to the mid-third century AD, whose exact chronological phases are difficult to reconstruct.874 The octagonal Room V stands out among these earlier structures (see fig.9.26), which include a furnace (Z), a heated room with hypocaust (S), a short balcony with an apsidal corner where a few people could stand and enjoy the view of the sea (P1). Room V was oriented north-south (while Room I will be oriented east- west), and communicated with Room M through a rectilinear small corridor.875 It was heated and featured a small semicircular alcove that leaned against its eastern wall whose use remains unknown (fig.9.27 ).876

9.27. Alcove on the eastern wall of Room V.877

The villa’s excavators have hypothesized that Room V and this sector of the villa was a bathing facility in the mid-third century AD. Yet, no plumbing arrangement has been uncovered and it is possible that the rooms were heated simply to permit their use even during the winter season. What was the use of this room in the mid-third century AD? Some hints can be offered by its

874 Guida, pp.101 ff.; figg.154-155. 875 Ibidem, p. 76, fig.102. 876 Guida, p. 81, fig. 111. On the adoption of polygonal rooms in Late Antique residential architecture, see SCAGLIARINI CORLAITA, “Gli ambienti poligonali nell’architettura residenziale tardoantica,” in XLII Corso di Cultura sull’arte ravennate e bizantina, Seminario Internazionale sul tema “Ricerche di Archeologia Cristiana e Bizantina,” Ravenna 1995, pp.837-873. 877 After Guida, p. 81. 238

octagonal design. The octagonal plan of Room V may in fact carry a Christian symbolism. For the early Christians the number eight symbolized Christ’s resurrection. According to the Bible, Jesus arose from the dead one day after the Sabbath, or seventh day, which early Christians counted as the eighth day (Sunday). This was the first day of the week and the day of the New Creation just as the old Creation also began on what is the first day of the week.878 Furthermore, the octagon was closely related to the circle. Gregory of (Opera, VIII, 2) described it as ‘a circle with eight angles.’ Thus, the octagon, for the Christians, continued the funerary association of the round plan.879 W. M. Bedard notes the similarity of plan between the octagonal baptistery at Salona (Dalmatia) and the Mausoleum of Diocletian at Split; the latter may have influenced the former.880 This resemblance suggests that there is a close link between tombs, martyria and baptisteries. For the early Christian there was an intimate connection between the resurrection of the body commemorated inside the mausoleums, and the resurrection of the neophyte celebrated inside the baptisteries.881 As Christianity spread, octagonal-shaped churches were built to reflect the symbolism of the octagon. The emperor Constantine, around 315 AD, commissioned the Lateran Baptistery in Rome (fig.9.28), which is composed of two octagonal rings.882

878 On the octagonal plan as a symbol of the resurrection, see DOLEGER, F.J., “Zur Simbolik des altchristlichen Taufhauses,” AC 4 (1934), pp.153-187. 879 BEDARD, W.M, The Symbolism of the Baptismal Font in Early Christian Thought, Washington 1951, p.16. 880 Ibidem, p.16, note 4. 881 As Bedard points out further evidence of the relation of tomb and baptistery includes the burials in the Aria baptistery at Ravenna; the baptism of Severus of Antioch in the Martyrium of St. Leontius; the addition of a baptistery to the Martyrium of Babylas at Antioch. BEDARD 1951, p.17, note 5. 882 The Lateran baptistery was restored by pope Xistus III (432-440). See ROGERS, C.F., Baptism and Archaeology, Oxford Clarendon 2006, pp.336-337. 239

9.28. The Lateran Baptistery in Rome (315 AD).883

In 327 AD, Constantine built at Antioch, in present-day Turkey, the ‘Golden Octagon.’ This octagonal building adjoined the Imperial palace on the Orontes Island in the centre of the city and was destroyed by an earthquake in 526. No trace has survived, but Eusebius, as well as later writers, provides a description of it.884

883 After http://en.wikipedia.org/wiki/Lateran_Baptistery 884 KRAUTHEIMER 1986, pp.79 ff. Eusebius, Vita Constantini, III, 50. 240

9.29. The Church of the Nativity in Bethlehem (333 AD).885

Octagonal buildings became common also in the Holy Land in the fourth century AD. In about 333 AD, Constantine commissioned the Church of the Nativity in Bethlehem, the place where Christian tradition holds Jesus was born. This church originally had an octagon at the head of the nave (fig. 9.29). It was destroyed during a Samaritan riot in 529, but was later rebuilt.886 At Capharnaum (Israel), an eight-sided room was found directly over ‘St. Peter’s House.’ It was identified as a fourth century house-church, which also contained a semicircular apse on the east side where a small baptistery was placed.887 Looking at these religious octagonal buildings, it is legitimate to suggest that the octagonal plan of Room V expressed a Christian function versus merely being a common design.888 Before the Villa at Palazzi di Casignana became monumental in the fourth century AD, Room V laid in noticeable position within the eastern sector of the villa. It was accessible

885 After KRAUTHEIMER 1986, fig.26. 886 KRAUTHEIMER 1986, p.61, fig.26. 887 CORBO, V.C., The house of St. Peter at Capharnaum; a preliminary report of the first two campaigns of excavations, April 16-June 19, Sept. 12-Nov. 26, 1968, Jerusalem, 1969. 888 The octagonal plan was not a novelty in the ancient and Roman architecture. See, for instance, the Octagonal Dining Room in Nero’s Domus Aurea. But in the fourth century AD developed a trend for octagonal building detached from or loosely attached to a main palace or church, which were noticeable from the outside. On this subject and several examples of octagonal buildings dated to this period, see the contribution by O. Brandt “Il Battistero Lateranense dell’imperatore Costantino e l’architettura contemporanea: come si crea un’architettura battesimale Cristian?” in FLEISCHER, J., LUND, J., NIELSEN, M. (ed.), Late Antiquity: art in context, Copenaghen 2001. 241

from corridor M and oriented north-south. When seen from outside, this room was prominent, having a roof that was likely made of timber and featured a pyramidal shape or a dome. According to M. Todd, octagonal rooms inside the British fourth century villas at Holcombe and Lufton were likely used for Christian baptism, even if they have been usually interpreted as bath-suites.889 In his contribution, Todd points out that octagonal structures are included in several Roman villas in the western provinces. A villa near Faro, in southern Portugal, contained a large octagon which had no evident links with a bathing facility.890 The villa at Lalonquette in Aquitania included an octagonal room adjacent to a baths suite floored by a mosaic featuring a central emblema. Based mainly on the architectural remains of these polygonal rooms, Todd suggests that they were used for Christian baptism. In regards to the fourth century Villa at Lalonquette,891 he writes:

‘The villas of the wealthy landed class of late Gallo-Roman society will certainly have witnessed baptism, but any relevant buildings or installations will not be easily detectable by structural remains alone.’

He also mentions the existence of several octagonal baptisteries, whose typology originated in Italy spread in Gaul (at Frejus, Marseilles, Aix and Riez) and in the Balkan provinces within Christian contexts.892 As C. F. Rogers points out, evidence shows that sometimes baths were converted into baptisteries.893 Thus, it is difficult for archaeologists to distinguish a thermal facility from a baptistery unless specific evidence emerges, such as an altar or an inscription that help to determine the function of space. However, J. Beckwith points out that there is no evidence of fixed altars before the fifth century, even in Rome. He states that portable altars must have been the general rule in early Christian times, and refers to a text by Minicius Felix stating that the Christians had no churches or altars. Also Arnobius, who wrote towards the end of the third and the beginning of the fourth century, stated that Christians had no temples, statues, nor effigies of their divinity, no altars on which to offer sacrifices, incense, libation and sacred meals. 894 The information given by these sources seem to justify the absence of material evidence in the majority of the early Christian contexts. Following Beckwith’s argument, if archaeologists do not find altars, objects of cult, statues, or any other liturgical data within an early Christian context, this is because early Christian worshippers did not use such furnishings. Christian communities, in fact, did not make any attempt to develop a monumental architecture and apparatus of their own before the reign of Constantine, and the Christian Church reached its full maturity in the fifth century AD.895

889 TODD M., “Baths or Baptisteries? Holcombe, Lufton and Their Analogues,” Oxford Journal of Archaeology 24 (3) 2005, pp. 307-311. Idem 1978 for a survey on Roman British villas. 890 ALARCAO 1988. 891 BALMELLE 2001, p.129, fig.52. 892 TODD 2005. 893 In the West of the Empire the octagonal and circular form prevailed, whereas in the East the small square or circular and sometimes the Greek cross. Se ROGERS, C.F., Baptism and Archaeology, Oxford Clarendon 1903, p.350. 894 BECKWITH, J., Early Christian and Byzantine Art, 1970, p.14. Marcus Minucius Felix was one of the earliest of the Latin apologists for Christianity. He wrote approximately between 150-270 AD. Arnobius of Sicca was also an early Christian apologist, during the reign of Diocletian (284 – 305). 895 BECKWITH 1970, p.14. 242

In short, it is possible that the octagonal Room V at Palazzi di Casignana served a Christian function in mid-third century AD, perhaps connected to the interesting small semicircular alcove that leaned against its eastern wall (fig.5.27). The absence of water provision weakens the possibility that Room V was part of a full-blown bath complex. If Room V had some cultic association, then it is very likely that the subsequent Room I built on top of Room V, continued to carry a religious function, namely that of a Christian chapel. With regard to Asia Minor, H. Leclercq writes: “dès le IVe siècle, L’église en forme de croix, bâtie sur plan octogonal et couronnée d’une coupole a tambour, était une disposition tout à fait usuelle et courante (The church cross-shaped built on an octagonal plan and crowned with a cupola drum was usual and ordinary).”896 If Room V was used for religious purposes, then Villa at Palazzi di Casignana may have hosted Christian activity already in mid-third century AD.

b) The cruciform plan of Room I

When the villa was renovated in the fourth century AD, the cruciform design of Room I was preferred by the villa’s owner perhaps because it fit symbolic reference to the Christian cross. In response to the criticism against the interpretation of architecture in terms of symbolism, Krautheimer explains that any building in the ancient world was “‘meant to convey a meaning which transcends the visual pattern of the structure.”897 There was a correspondence between the architectural design and its content. Each pattern or plan had its meaning. Krautheimer rightly underlines that:

“Buildings in antiquity, as today, did use after all a vocabulary expressive of their function and their place in a social or religious hierarchy. To deny this means closing one’s eyes to an element integral to any architecture.”898

Landulfus the Elder writes: Condidit Ambrosius templum, Dominoque sacravit nomino Apostolico, munere, reliquiis. Forma crucis templum est, templum victoria Christi, sacra triumphalis signat imago locum (Historia Medilanensis 1:7).899 This text makes it clear that the cruciform plan commemorates the victory of Christ and the cross is described as the triumphal image. In architecture, as Roger points out, “the cross-shaped plan evoked primarily Christ’s victory, not as the cross does today, Christ’s death or simply the Christian religion.”900 Rogers goes a step further and proposes that the cruciform shape equates the idea of Christ’s victory over death, which is the resurrection. At the Villa of Palazzi di Casignana, before Area D was renovated in the fourth century, the octagonal room V was orientated north-west. But when Area D was renovated in the fourth century, Room I was orientated east-west. This is the typical orientation of early Christian churches. In the five first basilicas built by Constantine, the apse of the church was at the west end (as for Room I at Palazzi di Casignana) and the priest, standing behind the altar, faced the

896 ROGERS 2006, p.38; H. Leclercq, “Eglises,” in DACL, 4, p.320. 897 KRAUTHEIMER, R., “Introduction to an ‘Iconography’ of Mediaeval Architecture,’ in Journal of the Warburg and Courtauld Institutes, V, 1942, pp. 1-33, especially p. 20. 898 KRAUTHEIMER 1986, p.491. 899 ROGERS 2006, p.39. 900 Ibidem. 243

east. This orientation probably derived from that of the church of the Holy Sepulchre at Jerusalem and the church at Bethlehem. Three fourths of the early churches in Rome followed this orientation, but in many churches it was reversed at a later date.901 It is noteworthy that the majority of the bodies buried on the villa’s archaeological area are orientated in the same direction as Room I (burials on the villa’s site will be discussed below on section d. According to Krautheimer, the prototype of cruciform rooms derives from religious architecture.902 The fourth century Church of the Holy Apostles in Constantinople was cross- shaped. The building has not left any trace but Eusebius describes it shortly after it had been built by Constantine:

“It was cross-shaped, the entrance arm perhaps slightly elongated. Gilded and coffered ceilings covered all four arms, but it remains to be decided whether they were all aisleless or divided into nave and aisles. The walls were covered with marble revetment...”903

Dozens of churches during the later fourth and the early fifth centuries copied the plan of Constantine’s Apostoleion. They are found, for instance, in Milan and Ravenna, at Ephesus and Antioch, and in the Holy Land.904

901 According to Tertullian the Christians of his time were believed to be votaries of the sun. This supposition, he adds, doubtless arose from the Christian practice of turning to the east when praying (Apol., c. xvi). Speaking of churches the same writer tells us that the homes “of our dove,” as he terms them, are always in “high and open places, facing the light” (Adv. Val., c. iii), and the Apostolic Constitutions (third to fifth century) prescribe that church edifices should be erected with their “heads” towards the East (Const. Apost., II, 7). See “Orientation of Churches,” in The Catholic Encyclopaedia http://www.newadvent.org/cathen/11305a.htm. 902 KRAUTHEIMER 1986, pp. 76-77. 903 Ibidem, pp.72-73. 904 Ibidem. 244

9.30. Villa of Fortunatus, Spain (late 4th c. AD).905

Going back to Palazzi di Casignana, a remarkable similarity with Room I and its flanking Rooms H and L is encountered in the late fourth century AD cruciform chapel inside the Villa of Fortunatus in Huesca, Spain (fig. 9.30: the chapel is marked on the left lower corner of the picture). This was built over an earlier three-aisled triclinium.906 Perhaps it was the basilica plan of the older dining room that inspired the choice for the chapel, which also contained a pseudo- crypt that was accessible by three narrow stairs and protected by rough stone screens. The pseudo-crypt imitated in small scale the crypts of the larger churches and probably held a

905 After BROGIOLO-CHAVARRIA ARNAU 2004, fig.65. 906 PALOL, P. “La arqueología cristiana en la Hispania romana y visigoda. Descubrimientos recientes y nuevos puntos de vista,” Actes du XI congres international d’archeologie chretienne,vol.2, Lyon 1989. 245

portable reliquary.907 Christian ownership of the residence is attested by a floor mosaic in one of the southern rooms of the complex. The mosaic features a floral and animal motif in which is inserted the name Fortunatus divided by the Christian symbol of a chrismon.908 A second example of cross design, very similar to Room I at Palazzi di Casignana, is attested in the Archbishop Chapel of Ravenna (fig. 9.31) built between the fifth and the sixth century AD, which was first dedicated to Christ and in a later phase to St. Andrew.909

9.31. Ravenna: The Archbishop Chapel (5th-6th century AD).910

907 BOWES 2005, p.202 908 A chrismon is a Christian symbol made of the superimposed Greek letters Chi (X) and Rho (P). These are the first two letters of the Greek word Χριστός (Christ). The Ρ is formed to look like a shepherd's crook and the Χ like a cross, symbolizing Jesus Christ as Good Shepherd of his flock, the Christian Church. The chrismon is also called labarum, which is the name of an imperial standard, where the chi rho was employed. 909 MACKIE, G.V., Early Christian chapels in the west: decoration, function and patronage, Toronto 2003, pp.104- 115. 910 After http://www.riminibeach.it/visitare/cappella-arcivescovile. 246

9.32. Small church at Caričin Grad (6th c. AD).911

In the southern part of the former Yugoslavia, at Caričin Grad, a sixth century small church was vaulted, cross-shaped and flanked by two smaller rooms, resembling very closely Room I at Palazzi di Casignana (fig.9.32).912 A similar plan was typical of Roman mausolea that have been found all over the Roman Empire.913 However, the cruciform plan was employed in aristocratic residences not only in spaces associated with Christian cult. A close parallel with Room I at Villa Palazzi di Casignana can be found in Tunisia inside the House of the Hunt in Bulla Regia dated to the fourth century AD.914 There, the cross-shaped Room N1 features an apsidal space that

911 KRAUTHEIMER 1986, p. 287, fig. B. 912 Ibidem. 913 Ibidem. 914 THÉBERT 1972; BESCHACUACH, HANOUNE, THÉBERT 1977; THÉBERT 1993, p.254-256, and 260-261; SODINI 1995, p.184; BALDINI LIPPOLIS p. 166, Bulla Regia 3. 247 likely contained a stibadium couch, a space that indicates that this room may have been used also as a triclinium. In contrast to Room I of Villa at Palazzi di Casignana, Room N1 at the House of the Hunt is inserted within the internal layout but not located in prominent position right at the entrance of the residence on the most monumental side of the villa. Also at Antioch-Daphne two rooms are cross-shaped.

9.33. Aquitania: cruciform rooms h-k.915

915After BALMELLE 2001, I, p.165, fig. 63. 248

Other examples of cruciform design recur in some fourth century villas of Aquitania Novempopulana in spaces that have been interpreted as reception rooms (fig.9.33);916and in two rooms of the Yaqto Complex of Antiochia (Turkey) dated to the third century AD, which feature mosaic decorations that have been interpreted as reception rooms.917 These examples show that during Late Antiquity, and in several regions of the Empire, there are many similarities between early Christian buildings and Roman villa architecture. For instance, apsidal halls that we find in many late Roman villas were taken over to form the simplest early Christian churches. Triconchs recur in baptisteries, martyria and mausoleums.918 Early Christian architecture, it seems, borrowed elements from the late Roman villa plan. As noted earlier, Room I opens onto the seafront Portico P and is not inserted within the villa’s plan in a secluded location. This seems to indicate a more public function of Room I rather than a private and intimate one. Religious spaces inside villas that are entered directly through the residence itself, as in Lullingstone (fig. 9.34), seem to have had a more private character and sheltered the devotion of the owner and his family.919

9.34 Villa at Lullinstone, England (400 AD). The small chapel is indicated by the shaded areas.920

916 The villas in Aquitania Novempopulana that feature rooms with a cruciform plan are those at Saint Sever, Jurancon Pont d’Oly, Lalonquette and Montcaret, see BALMELLE 2001, p.165 (fig.63h-k). 917LASSUS 1938; idem 1980, p. 362; BALDINI LIPPOLIS p.134, Antiochia 7. The domus includes at least 41 rooms and underwent two main phases of construction. The first one is dated to the third century AD, the second one to the fifth century. 918 KRAUTHEIMER 1986. 919 For the Villa at Lullinstone see MEATES 1987. 920 After BURRUS 2005, p. 202, fig.8.4. 249

On the contrary, the long apsidal room (20 m.) inside the Sicilian villa at Patti Marina, lies outside the main circulation axis (fig. 9.35). For this reason, Wilson believes that it may have been an “estate-church, added in the second half of the fourth century when the owner had likely embraced Christianity.”921

9.35. Sicily: Villa at Patti Marina.922

L. Pietri argues that in Late Antique Sicily the wealthy landowners who converted to Christianity promoted the construction of religious structures inside their villas as a way to become ‘eternal.’923 This scholar based his interpretation mainly on the letters written by the popes Gelasius (492-496), Pelagius I (556-561) and Gregory the Great (492-496), and concludes that besides Rome, Southern Italy and Sicily is the area where the largest numbers of private religious foundations may be found.924 Noteworthy is also the transformation of late Roman villas into monasteries although this passage is not always detectable archaeologically. We know, for instance, that Cassiodorus founded a monastery on his family villa-estate at the native Scolacium in Bruttii in 554 AD.925 It was called Vivariensis sive Castellensis and housed monks who were devoted to eremitism but also to fish-rearing, as the first name of his foundation indicates.926

921WILSON 1990, p.206. 922 After WILSON 1983, fig.48. 923 PIETRI 2002, p. 253. 924 Ibidem; SFAMENI 2006, pp. 268 ff. 925 Cassiodorus, Institutiones, 1,29,3. 926 See also FARIOLI 1982, pp.674-676; MINUTO 1990, pp.309-314. 250

9.36. Calabria: The Church of San Martino in S. Martino di Copanello.927

The triconch of the Church of San Martino in S. Martino di Copanello (fig. 9.36), according to P. Courcelle, should be identified with the core of Cassiodorus’ monastery.928 Farioli argues that the same triconch in earlier times belonged to Cassiodorus’ family’s villa that preceded the monastery.929 At Tauriana, near Cosenza, in northern Bruttii there is evidence of a monumental nimphaeum belonging to a late imperial villa that between the fourth and fifth century was transformed into the monastic complex of S. Fantino il Vecchio.930

927 After FIACCADORI 1994, p.723, figg.14-15. 928 COURCELLE 1938. 929 FARIOLI 1982; ZINZI 1998, p.47. 930 COSTABILE F., “Il ninfeo romano ed il complesso monastico di S. Fantino a Taurianum,” in Klearchos, 1976, pp.83-120; FIACCADORI 1994, pp.728-729. 251

c) The Transformations of Room D and E in the Fourth Century AD.

9.37. Villa at Palazzi di Casignana: Room D with staircase. Room E is in the background.931

931 After Guida, p.81, fig.110. 252

At Palazzi di Casignana, the location of a well behind the apse of Room I, the storage of water in Room E, and the presence of three steps in Room D (fig.9.37) may provide further evidence that the apsidal Room I was a chapel. During the latest phase of Room I, around the middle of fourth century AD, significant modifications were made to the adjacent Room D and Room E. They were transformed into water containers. The height of the walls of both rooms is not known since the upper structures are not longer extant. No plumbing system for the provision of water was ever constructed; only a well was built behind the apse of Room I. This section will describe the adjustments made in the fourth century AD to the northern sector of the villa and attempt to give an explanation for their construction. Room D, which was accessible through Corridor G, received a three-step staircase. Only two steps are preserved while the third one is badly damaged (fig.5.37). The stair was built against a wall that was erected to close up the only entrance to the adjacent room, Room E. This last room was transformed into a water cistern and its walls were coated with hydraulic plaster. To do this, the entrance to Room E located on the western wall of Room D was walled. Archaeologists of the villa suggest that Room E and Room D were unroofed and used as impluvia to collect rain. Yet, the absence of a roof in both rooms not only would seem incompatible with the reconstruction of the eastern façade proposed by de Nittis, which included a second story and flanking towers (for a discussion on the eastern façade reconstruction see Chapter 4 of this study), but it would have been discordant with the luxurious character of this part of the villa. Room E had two doors before it was transformed into a water cistern suggesting that it was a roofed room. A restoration of the layout and roofing system together with an interpretation of the function of this sector of the villa still lie in the dark due to the poor state of preservation of the structures and the damage caused by the construction of road S.S. 106. For now, it seems possible to state that since the walls of both Rooms D and E were not perfectly orthogonal, because they were built in different phases and with different techniques over time, it would have been difficult for these walls to support a very high, heavy, and monumental roof. In interpreting the purpose of the staircase in Room D, archaeologists have hypothesized that it was used to reach a second story located on top of the adjacent Room B, which was one of the most elegant rooms of the villa featuring a beautiful figural mosaic representing the Four Seasons. Yet, if a second story existed, as they specify, Room B would had to have a very low ceiling since its walls were not perfectly orthogonal and also not solid enough to support a high ceiling. Thus, a low ceiling in Room B was unlikely because it would have spoiled the opulence and the importance of Room B in this prominent section of the villa. The second alternative, according to the archaeologists, is that the stair of Room D was used to reach a mezzanine or to inspect part of Room D that functioned as an impluvium. What if the staircase of Room D was used instead to inspect and reach Room E? Room E may have been roofed rather than open. Since both entrances of Room E had been walled when Room E was transformed into a water cistern, the staircase of Room D would have been the only way to reach Room E and to fill it with water. It is not possible to say how high the walls of Room E were; they may have been high enough, probably over a meter, to contain a good amount of water but not necessarily as high as to reach the roof. To check the water or access Room E there seems to be no better way than coming from above using the stair located in Room D, for there were no other openings and no water conduits furnishing water to this room. The task of filling Room E with water was likely left to servants that had to carry water in buckets.

253

They may have drawn the water from the well located behind the apse of Room I (fig.9.38), which perhaps was furnished by the same aqueduct on the villa’s baths. 932

9.38. Villa at Palazzi di Casignana. Well behind the apse of Room I.933

932 Water supply even in the wealthiest residences was confined to the ground floor. All ancient texts indicate absence of installation that conveyed spring water to the upper level. No building that has been so far excavated in Rome or Ostia, for instance, reveal trace of rising columns that may have carried water to the upper stories. For this reason, water carriers (aquarii) were vital figures within the domestic life of the Romans to the point that they were inherited as part of the building itself by the heir or successors. In the Satires of Juvenal they are described as the scum of the slave population. (Juvenal, 6, 332). CARCOPINO 1940, p.39 note 74. Villa at Palazzi di Casignana’s owners may have used aquarii for the provision of water inside the baths and the residential area of the villa. 933 After Guida, p. 103, fig.157. 254

9.39. Eastern area of the Villa at Casignana.934

Room E previously had two entrances, on the northern and eastern side (see red parts on fig.9.39). The excavators’ hypothesis that it was transformed into a water reservoir in order to collect rain water by creating two impluvia from spaces that were used previously as regular rooms seems odd. Especially considering that during the same phase of the villa’s renovation the provision of water was abundant enough to furnish not one but two whole suites of baths! What motivated the villa’s owner to devote Room D and Room E to function as water cistern? Three different explanations are perhaps possible. 1) water was used for the daily needs of the villa’s residents and also during dinners and receptions in this area of the villa. This seems unlikely. Certainly, the need of water in this part of the villa was not substantial enough to justify the high expenditure and the technical enterprise of constructing a water system that supplied running water on a daily basis. 2) A second possible reason for transforming Room E into a water cistern might have been to insure there was enough water in case of fire. The praetorian prefect Paulus, in giving instructions to the praefectus vigilum (the commandant of the Roman firemen) reminded him that it was part of his duty to warn tenants always to keep water ready in their rooms to check an outbreak.935 Yet, the Villa at Palazzi di Casignana was only a few meters away from the sea which provided water in great abundance. Dismantling Room E in order to create an impluvium for collecting rain water to put out any possible fire seems a far-fetched explanation. 3) Another possible use of water might be for the Christian liturgy. If one assumes that Room I in fourth century AD may have hosted Christian rituals, the water contained in Room E could have been used for baptism or sacred services. It may be speculated that Room E was roofed and had low walls that formed a tub used as a piscina for baptismal immersion. The tub would have

934 After Guida, p.80. 935 “ut aquam unusquisque inquilinus in cenaculo habeat iubetur admonere.” Digest I 15, 3, 5. 255 been accessed through the three steps of Room D, whose number three may even carry a symbolic meaning. The neophytes passed by steps, in going down into the water; to the ascent and descent, as well as to the number of steps this involved, there was often attached a mystical significance (Isidore of Seville, De divin. off., II, xxv).936 Baptisms could be performed only at Easter and at the Pentecost of the liturgical year.937 Therefore, Room E did not need to be filled with water all year long. The infrequent use would explain why Room E was not furnished with water conduits that were costly to build and elaborate plumbing. Water was, however, an essential component in Indoor baptism (Acts 9:18; 16:33). C. Rogers points out that:

“Before the conversion of Constantine Christian worship could only be carried on in private houses (domus ecclesiae), where the wealth of the owner formed a protection, or in the catacombs…”938

Baptism in private houses lasted until it was forbidden in 527 at the Council of Dovin in Armenia. The same prohibition was reinstated at the Council of Constantinople in 692, unless the bishop granted permission. 939There were no

No parallels can be provided here for Room E but it is important to keep in mind that the practice of baptism within the private sphere did not require any specific architecture or apparatus, at least until Christianity reached full maturity in the fifth century AD. All the new converted needed in order to be baptized was water. Tertullian wrote a panegyric on water, which is the source of life, a material “always perfect, joyous, and simple and pure” (Bapt.3, and 4).

W. Bedard states:

“The adoption of the quadrilateral [baptistery], which seems to have been the shape of the earliest baptisteries in all areas, was probably due to three influences. First, when baptism ceased to be administered in rivers or pools and a baptistery was arranged in the house-church, it naturally had the shape of an ordinary room, i.e. it was quadrilateral, and as such it has been noted at Dura Europos …

The baptistery therefore continued as a side-room, quadrilateral in shape; …what took place in the baptistery resembled closely what took place in the public baths, the influence of the Roman frigidaria, which was often a square or rectangular room, sometimes with an apse, must be acknowledged.”940

In another passage Bedard writes:

936 ROGERS, C.F., Baptism and Christian Archaeology, 1903, pp. 314-315. 937 Pope Leo I Magnus (440–461) exhorted the bishops of Sicily to celebrate baptisms exclusively on Easter and Pentecost condemning the irrational novelty (irrationalibis novitas) of administering the sacrament even on the Epiphany. Epistulae, 16: PL 54, pp.695-704 938 ROGERS 1903, p.314. 939 Ibidem, p. 315. 940 BEDARD 1951, p.13. 256

“From New Testament times baptism was understood in terms of death and burial. Jesus himself had seen his crucifixion as a baptism (Luke XII. 50), and St. Paul spoke of being ‘buried …with him through baptism into death’ (Rom. VI.4). The font and also the room that contained it were consequently regarded as a tomb… Ambulatories were a normal feature of pagan funerary monuments, e.g. the mausoleum near the Porta San Giovanni at Rome, and as such were reproduced in the baptismal font wherein the Christian died and rose again with his Lord.”941

Thus, quadrilateral baptisteries were built because of their association with tombs, which indeed were squared in many areas of the Roman world, including Italy, the East, North Africa, and Palestine.942 At Palazzi di Casignana, the quadrilateral Room E may have been used as a baptistery and Corridor G, the ambitus (fig.9.39), which was narrow and adjacent to Rooms I, D and E, likely had a ritual and ‘ceremonial’ function connected to Room I.

941 BEDARD 1951, pp.14-15. 942 Ibidem. 257

Burials near the Villa at Palazzi di Casignana.

The hypothesis that Room I functioned as a chapel could be supported through the association with burials uncovered near and inside the villa. The act of burying has strong religious components. Placing tombs near a villa that included a religious space might have been a natural choice for the local community. The construction of a church and a cemetery was, for instance, contemporary at the fourth century Portuguese villa at São Cucufate, whereas at the Spanish Villa of Fortunatus (4th c.) the placement of graves preceded the construction of a private chapel.943

9.40. The site of the Villa at Palazzi di Casignana and its necropolises.944

The two necropolises found at the north-east and south area of the Villa at Palazzi di Casignana (fig. 9.40) have been vaguely dated to the ‘imperial phases’ of the site leaving unspecified how many were contemporary to the villa’s inhabitation or of a later date. 945 The north-east necropolis was only partially explored in the 1970s, while the southern graveyard has been more thoroughly investigated. The latter contained 91 tombs of which 23 have been thoroughly examined by the excavators. No graves included goods except for one grave (2001/Tomb 2), which contained two small, damaged coins and one lamp of African sigillata dated to the mid-

943 RIPOLL-ARCE 2000, p. 94. 944 After Guida, p.35. 945 For a detailed description of the two necropolises see Guida, pp. 89-90. 258

second century AD.946 It has been suggested that the southern necropolis was used for the entire life of the villa. Perhaps a coastal road that connected Locri with Rhegium ran nearby. This road has not been yet identified by the archaeologists, but its existence is attested by ancient sources.947 More recently, other tombs have emerged near the eastern sector of the villa where Room I is located. They are not mentioned in the archaeological guide of the villa as their discovery occurred after the publication of the book.948 Even inside the villa, in the gallery of the praefurnia located in the western baths, inside the corridor of access to frigidarium 12, and outside Portico 17, several burials were found including that of an infant placed inside a amphora Keay 52 discovered near the nymphaeum.949 Who were the people buried in the necropolises of the villa? Why did their families choose this site? What type of community lived around the villa? What was the social background and religious beliefs of the people buried on the site, especially in the consideration that Locri’s territory was already largely Christian during that time? These questions are not easily answered due to the premature level of archaeological research at the site of Villa at Palazzi di Casignana. Very likely the graves belonged to the population that gravitated around the villa’s estate. The choice of this area may have been in great part of religious character.950 All of the burials show the practice of inhumation and the majority is oriented east-west. Only a few graves contain grave goods. These characteristics are typical of Christian funerary practice. The graves so far analyzed by the archaeologists range in date from the end of the third century to the seventh century AD, when Christianity was largely diffused across Bruttii, and the territory of Locri.951 The presence of several buildings around the Villa at Palazzi di Casignana strongly suggests that the area was densely populated. It is highly possible that the villa and its owner played a religious role within the rural community when the villa was still inhabited. After the villa fell in disuse, graves continued to be placed near the building, and also inside the praefurnia of the baths, or in the frigidarium of the Western Baths (Room 12).952 The attraction of the site as burial ground apparently continued even after the villa was abandoned. During Late Antiquity, burials were being placed in both open and roofed spaces inside late Roman villas or leaned against its external walls, at times even when the villas were still occupied. The position of burials near private villas indicates that the elites of the third and fourth centuries may have evolved their rules and belief system. These graves reveal a desire to keep their dead close to and at times even inside their homes. There is little or no evidence of this phenomenon connected to Roman villas from earlier periods, with the exception of infant

946 Ibidem. 947 A synthesis on the road network of Roman Bruttii is provided by GIVIGLIANO 1994, 1989, 1986 and 1981. See also NOYÉ 1988, pp.69-70. 948 I personally saw the excavation of these tombs in Summer 2008. 949 BARELLO- CARDOSA 1991, pp.677-679, figg. 3.1 to 3.3, and 4; Guida, pp.91-93. 950 J. Percival in the 1970s had devoted an entire chapter to the practice of burying the dead in proximity or inside late Roman villas. He took into account especially the archaeological evidence from the north-western provinces. PERCIVAL 1976, pp. 183-199. 951 By the time of Hadrian, cremation fell in disuse and by the fourth century Macrobius refers to cremation as a custom that belonged to the past. The reasons for the switch from cremation to burial has been attributed largely to Christianity. Yet, it is important to keep in mind that evidence from inscriptions and tombstones across the Empire shows that for the most part Christians did not separate themselves from non-Christians when burying their dead. According to E. Rebillard it would not be until the early Middle Ages that the Church gained control over burial practices and that “Christian cemeteries” became common REBILLARD, E., Religion et Sépulture: L'Église, les vivants et les morts dans l'Antiquité tardive, 2003; idem 2009. 952 Guida, p.89. 259

burials, which have been found in several dwellings of Italy, Gaul and Britain.953 The pre- existing boundary separating the living and the dead significantly changed during Late Antiquity. Burying the dead near the living became accepted even inside urban boundaries.954 Previously, the dead were typically buried away from living spaces and nearly everyone was buried beyond the limits of the city or pomoerium. A law from the Twelve Tables in fact ordered: hominem mortuum in urbe ne sepelito neve urito (translation: none is to bury or burn a corpse in the city).955 Several recent archaeological discoveries in the Suburbium of Rome show that the custom of burying the dead near inhabited villas started between the third and fourth century AD.956 It is unclear if this custom was determined by practical reasons, such as the availability of a sheltered space and construction material; the necessity of sparing farming land for cemeteries; or if it was more dictated by a change in the relationship between the dead and the living. The latter hypothesis appears more likely. Christian beliefs in the afterlife may have played a role in the advent of this new practice. In fact, during the late Empire, a trend began inside aristocratic properties that blended Christian cult and dead commemoration. Sulpicius Severus tried to have the body of Martin of Tours to worship it in his property, but had to be content with that of Clarus, Martin’s disciple, which he kept in a private church (martyrium) on his estate together with other relics from the Holy Land.957 Owning Holy Land relics must have increased the status of his estate and monastery community. The partial or total occupation of abandoned villas and their estates by a burial ground is a widespread phenomenon, that started at the end of the third century AD and developed especially during the fifth and sixth centuries. In Northern Italy two examples of burials inside late imperial villas can be cited: the Villa at Desenzano and the Villa in Sirmione known as ‘Grotte di Catullo.’ The sectors A and B of the Villa at Desenzano revealed a large number of burials.958 The inhumations in sector A, found inside the rooms adjacent to the peristyle and the triconch, were placed over the mosaic floors or overlying the walls. They did not contain grave goods and were dated to the early Middle Age. In sector B, within the apse of a large hall (see fig. 5.19, Room 35), which is decorated with an opus sectile floor, an infant burial was carefully placed in a elongated pentagon structure. This ‘privileged’ burial and the fact that the opus sectile floor was laid on top of it suggested that this space was used for worship, perhaps as a private oratory,

953 Cremation, which was predominant in the first and second century AD was denied to infants (those who died ‘before they cut their first teeth’), see Pliny, NH, 7, 15.72; Juvenal, Satirae 15, 138-140. See also the contribution by MOORE, A., The Roman house in Britain; Hearth and Home: The Burial of Romano-British Infants within Domestic Contexts, University of Southampton. 954 RIPOLL-ARCE 2000; CANTINO WATAGHIN 1999. See bibliography in BOWDEN-MACHADO 2006, pp.69-70. 955 Tabula X. See CRAWFORD, M., Roman Statutes, II, London 1996; HUMBERT, M. (ed.), Le Dodici Tavole. Dai Decemviri agli Umanisti, Pavia 2005. 956 SUBURBIUM 2000; DI GENNARO, F., GRIESBACH, J., “Le sepolture all’interno delle ville con particolare riferimento al territorio di Roma,” in PERGOLA, P. - SANTANGELI VALENZANI, R. - VOLPE, R., Suburbium. Il Suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno, Rome, 2003, pp. 123-166. For a discussion on the villas of the Suburbium see VOLPE 2000. See also SPERA 1999 on the landscape of the suburbium of Rome. Burying the dead inside the city walls of Rome became more common during the fifth and sixth centuries AD, see MENEGHINI, R., - SANTANGELI VALENZANI, R., “Il paesaggio urbano della tarda antichità”, in Aurea Roma (Catalogo della mostra), Roma 2000, pp.45-48. When this new custom emerged, it must have drastically collided with the documented institutional precedent laws that banned burials inside urban walls. 957BROWN, P., The cult of the saints, Chicago, 1981, p.31; BURRUS, V. (ed.), Late ancient Christianity. People's History of Christianity, Vol. 2, 1982, p.196. 958 BOLLA 1996, pp.51-70, especially pp.51-54. 260

during the late fourth and the fifth centuries.959 Inside the ‘Grotte di Catullo’ at Sirmione fifty burials were found. They are dated between the mid-fourth and the early fifth centuries with occasional later graves. The appearance of small hearths inside the villa throughout Late Antiquity indicate that part of the complex continued to be occupied while other sectors were used for burials. This evidence, which is encountered also in other sites all over the Empire, disproves the belief that the existence of burials inside a villa implied automatically its abandonment and the transfer of the population to a new site. 960The same can be said for the Villa at Palazzi di Casignana. In fact, the two necropolises found on the villa’s estate seem to have been used throughout the entire life of the complex and even after the villa was abandoned.961 Traces of hearths and ceramic dumps found in a few tubs of the baths dated to seventh century AD show that some sectors of the villa continued to be inhabited while burials were placed inside rooms that fell in disuse.962 A detailed study carried out in Normandy has detected two models of cemetery occupation during the Merovingian age (mid-fifth century to 751 AD).963 In the first, the graves are limited to a part of the residential buildings or to the bath complex; in the second, the burials are found in the pars rustica of the villa. Normally the burials, reaching as many as a hundred or so, follow the lines of the walls of the pre-existing structures. At the baths of the villa Gisay-la-Coudre (Beaumesnil, Eure), eight tombs were dated to late sixth or seventh centuries were found overlying a third-century destruction layer. It’s assumed that they were related to a possible use of the site as a church.964 In a recent article, in regards to early Christian burial basilicas in North Africa, A. M. Yasin writes:

“In comparison, however, with early Imperial tombs, which memorialized individuals primarily in terms of their status within the family and served as vehicles for the self- presentation of the household group, Early Christian burial basilicas in North Africa of the fourth to sixth centuries CE redefined the concept of family. By gathering the graves of unrelated Christians into a common space, these churches commemorated them with homogeneous memorials and expanded the commemorative audience to the entire local community of coreligionists.”965

So far, it is not possible to claim that the Villa at Palazzi di Casignana ever hosted Christian rituals, or that the placing of tombs around the villa was connected to a Christian chapel inside the building. Yet, it would seem natural for the local community to bury their dead near or in a location with religious significance. We may infer that the villa’s owner participated in development of the Christian faith in the region by providing a religious space for the worship of his family and of the rural community on his estate. As the study of K. Bowes has shown, in

959 SANNAZZARO, M., La cristianizzazione delle aree rurali della Lombardia (IV-VI sec.). Testimonianze scritte e materiali, 1990, pp.32-33. 960 PERCIVAL 1976, pp.184-185. A different view is held by RIPOLL – ARCE 2000, p.93. 961 Guida, p.89-90. 962 Guida, p.91. In some cases the grave cut was sufficiently deep to damage the underlying structures, as demonstrated by some rectangular holes visible on the floors. 963 LE MAHO 1994; RIPOLL- ARCE 2000, p.89. 964 RIPOLL-ARCE 2000, p.89. 965 YASIN, A.M., “Funerary Monuments and Collective Identity: From Roman Family to Christian Community”, The Art bulletin 87, no. 3, (2005), p. 433. 261

many provinces of the Western Empire, such as Hispania, the powerful landowners and their estates were the real engine of Christianization.966 One may assume the same for the owner of Villa at Palazzi di Casignana and his estate. If there is any indication that Villa at Palazzi di Casignana had a religious space, this would be represented by the apsidal-cruciform Room I on the eastern side of the villa. The necropolises of Palazzi di Casignana deserve a thorough examination. New findings could provide archaeologists and historians with powerful insights into the history and world of the villa and its inhabitants. Remarkably, they stand as a proof that during the late imperial phases of the villa “the world of the dead and the world of the living came closer and closer together.”967

966 BOWES 2005, pp. 204-205. 967 Ibidem, p.94

262

CHAPTER 10: CHRISTIANITY IN FOURTH CENTURY BRUTTII

Introduction

Evidence that Room I may have been a private or semi-private chapel can be seen in the presence of a diffused Christian community in the territory of Locri and in Bruttii in general. Constantine had officially adopted Christianity in 312 AD and by the end of the fourth century many wealthy and influential families across the Empire had embraced this religion. In 330 Locri became a Dioceses and it is around this time that Villa at Palazzi di Casignana was renovated, expanded with the additions of several rooms, including the seafront residential area D that included Room I. Literary texts show that Christianity was extensively spread across Bruttii in the fourth century AD.968 Between 313 and 319 AD, Constantine sent an imperial constitution to the governor of Lucania et Bruttii, Rufinus Octavianus (corrector Lucaniae et Bruttiiorum), with which the emperor absolved the clergy of this province from all duties (ab omnibus muneribus), so they could fully devote their time to the service and needs of the Church.969 This document proves that during the reign of Constantine, an ecclesiastic hierarchy was already established in Bruttii, since the clerici had to respond to the increasing demands of a large Christian community.970 This is confirmed by the discoveries in the provinces of Lucania et Bruttii of inscriptions with Christian content dated to the fourth century AD, some of which attest to the existence of bishops in both regions.971 In Locri and Rhegium, epigraphical texts and lamps dated to the fourth century AD name episcopi, clearly attesting that there was a mature and hierarchal Christian community organized in dioceses. In Gerace, a town located on the hills near Locri, and where the dioceses of Locri moved in the fifth century AD, a Christian inscription reports the name of an aristocratic man named Leporius who lived when Quintus Aurelius Symmachus was consul in 391 AD.972 According to G. Otranto, the Christianization of the southern Italian regions, including Campania, Apulia-Calabria, Lucania-Bruttii, and Sicilia was dominant by the end of the fourth century AD.973 Otranto argues that Christianity first spread along the coasts and then into the hinterlands, gradually developing in forms that may have varied from region to region. In all these southern Italian regions, initially, the Christian communities arose spontaneously and were

968 The Christianization of Calabria has been studied in detail by G. OTRANTO: “Cristianizzazione del territorio e rapporti col mondo bizantino,” in L’Italia Meridionale in etá tardoantica. Atti del trentottesimo convegno di studi sulla Magna Grecia, Taranto 2-6 Ottobre 1998, pp.66-113. 969 Codex Theodosianus, 16, 2, 2. 970 OTRANTO 1998, pp. 325-355. 971 An interesting inscription dated to the mid-fourth century AD was found in Blanda Iulia, the modern Tortora, on the border of Lucania and Bruttii, which refers to an Iulianus episcopus (ICI V, n.52, p.62). Another inscription dated to the same period comes from the territory of Copia-Thurii and refers to an archbishop named Marc/us or Marc/ellus (ICI V, n.48, p.57; see OTRANTO, op.cit. pp.356-357). In Metapontum a paleochristian basilica with an annexed baptistery both dated to the sixth century AD point to the presence in this town of a well established Christian community that scholars believe started in the third century: NOYÉ G., “Quelques observations sur l’evolution de l’habitat en Calabre du V au XI secle,” in Rivista di Studi Bizantini e Neoellenici n.s. 25 (35), 1988, pp.66, 83; OTRANTO 1998, pp. 79, 151-152. 972 CIL X 37; FIACCADORI 1994, p. 740. 973 In contrast, on the survival of traditional pagan cults, see MacMULLEN 1981 and 1997. 263

loosely organized.974 The evangelization started as early as the first century AD, whereas in northern Italy a widespread adoption of Christianity occurred only with the episcopate of St. Ambrose in Milan.975 The diffusion of Christianity across southern Italy is also documented in letters written by Pope Innocentius I (401-417)976 and Caelestinus I (422-432)977 to the Christian communities of those regions. These letters reveal a mature Christianization of Bruttii together with a strong interest of the aristocrats in becoming bishops.978 In the fourth century through the sixth century AD, several bishops from southern Italy, including a few from Bruttii, participated in Church councils and embassies to the East to defend the Roman Church orthodoxy against the rising heretical doctrines such as Aryanism. Southern bishops seem to have been firm opponents of heresies being aligned with the official religion. 979 There was a bilingual character of the early Christian community in Bruttii, particularly in the area between Locri and Rhegium, where inscriptions reveal Christians with Greek names.980 At Palazzi di Casignana, a Greek name is attested by a remarkable tomb uncovered north of the nymphaeum, which was orientated east-west. It contained two bodies, including that of Roupheinoo, a Greek name inscribed on the tomb slab. The tomb was placed on fallen structures of the nymphaeum suggesting a date subsequent to the abandonment of the villa.981

Christianity and the late Roman aristocracy

It is very likely that the rural population around Locri was largely Christian during the late phases of the Villa at Palazzi di Casignana. There is little architecture in the region identifiable as early Christian, but there must have been religious gatherings at this time, especially in the countryside.982 Rituals and religious beliefs have always constituted a driving force in rural life.983 It can be speculated that the dominus of villa at Palazzi di Casignana, being

974 OTRANTO 1998, pp. 69-113. 975 CRACCO RUGGINI 1961, p.68. 976 Epistula 39: PL 20, 606 977 Epistula 5: PL 50, 436. 978 GUARNIERI, C., “Note sull’elezione episcopale in Apulia all’inizio del V secolo”, in Vetera Christianourm, 17, 1980, pp.347-356;OTRANTO 1998, p. 90. 979 OTRANTO 1998, pp. 104 ff. 979 Ibidem. 980G. Rohlfs contends that the Greek language was continuously used in Bruttii until the Norman domination. See ROHLFS, G., Scavi linguistici nella Magna Grecia, Roma 1933, passim. Rohlfs also believes that Cassiodorus’ Vivarium was a Greek monastery, a theory that was disputed by P. Courcelle who proved instead the Latinity of Cassiodorus’ religious foundation. See COURCELLE, P., Les Lettres grecques en Occident, Paris 1948, pp.318 ff. The evangelization of Bruttii is anchored to the Eastern Empire. Magna Graecia had from the beginning of its history rooted relationships with the Greek world. See also PRONTERA, F., “L’immagine della Magna Grecia nella tradizione geografica antica,” in AA.VV., Magna Grecia. Il Mediterraneo, le metropoleis e la fondazione delle colonie, a cura di G. Pugliese Carratelli, Milano, 1985, p.9. 981 Guida, p.93. It is not specified if the two corpses were buried at the same time. If not, it is likely that the tomb had some type of grave marker that made it recognizable to the family members of the deceased when the second body was laid. 982 Besides the study of Otranto, on the diffusion of Christianity in Calabria, see: BARILLARO, E., Reliquie paleocristiane nella Locride, S. Giovanni di Gerace, Nossis, 1972; RUSSO, F., Introduzione del Cristianesimo nella Sibaritide, in Testimonianze cristiane antiche e altomedievali nella Sibaritide, pp.5-21; idem, Storia della Chiesa in Calabria, dalle origini al Concilio di Trento, Soveria Mannelli (Catanzaro), 1982, vol. 1, pp.37-157. 983 The boundaries of the estates, for example, were traced with markers that were consecrated with a ritual sacrifice; SICULUS FLACCUS, De condicionibus agrimensorum, in CLAVE-LEVEQUE, et Alii (ed.), 1993, pp. 46-70. 264

a leading figure in the area, created inside the largest and most monumental room of his villa (Room I) a religious space not only for his personal use but also to share it with the people on his land or the converted who lived in the area. Building a liturgical space would have proclaimed the importance of the villa’s owner and provided a way for the villa dominus to assert his influence and authority over a rural community that by now had become progressively Christian. Traditionally Roman historians have attributed the socio-economic changes that took place during Late Antiquity to the administrative reform of the provinces operated by Diocletian, which greatly affected the Italian peninsula during the later third and fourth centuries AD. More recent scholarship considers also the influence of early ecclesiastic institutions that during the third and fourth century were consolidating their strength and revitalizing society. The conversion of many Roman families created a new form of aggregation among the nobles and generated a change of their way of living.984 This has been confirmed by new archaeological discoveries in southern Italy, especially some remarkable ones in Apulia (Puglia) like the one at San Giusto, which revealed a late Roman rural establishment that includes a double basilica and a baptistery (fig.10.1).985

984 SALZMAN 2002. 985 VOLPE, G. et Alii, San Giusto. La villa, le ecclesiae, Lucera 1995-1997; VOLPE 1998, pp.223-329, especially pp. 294 ff. 265

10.1. San Giusto (Puglia): the site and its reconstruction.986

986 Reconstruction by F. Taccogna, L. Tedeschi. See VOLPE 1998, p.204, fig. 262. 266

The monumentalization and refurbishing of several late Roman villas, in some instances, may have been concomitant to the addition of sacred spaces. The addition of spaces for worship inside some villas must have contributed to their greater monumentality. K. Bowes looks at villas in which a space for cult has been identified, even with a low degree of certainty.987 These include the fourth century Portuguese villas at Milreu and Sao Cucufate,988 the Spanish villa at Carranque with its dome building989 and the villa at Pueblanueva with its eastern alcove.990 Bowes suggests that these villas may have been transformed into grandiose residences not simply for the personal aggrandizement of their owners but because they became places of worship. Their monumentality and their noticeable location within a particular territory signified aristocratic wealth and prestige but also the pietas of the dominus.991 Bowes states:

“The villa-temple, mausoleum or martyrium served its function not simply by acting as a space of cult or burial; in its monumentality and its careful topographic placement it also functioned as a language, transforming personal pietas and commemoration into markers of place and identity, thereby rendering them part of the broader estate apparatus.”992

The study of late Roman villas needs to look harder for evidence of Christian religious practice in these buildings. The scant evidence of credible literary sources for this period is a real obstacle in the reconstruction of early Christianity throughout the Empire. Yet, scholars can rely upon the rich hagiography of the early Middle Ages to shed valuable light on the earliest phases of Christianity. The rigid separation between different disciplines such as epigraphy, topography, literature, art history and archaeology has complicated the task of reconstructing an early Christian landscape in several areas of the Empire. Typically scholarly studies have focused on the main centres of Christianity such as Rome and Constantinople, without providing analysis of small, provincial communities and dioceses of the Empire. The architecture and decoration of late Roman villas and buildings in general may reflect a cultural background that is no longer fully pagan.993 The double set of baths, as those uncovered at Palazzi di Casignana, for instance, have been usually interpreted as a reflection of greater wealth, prestige or a means to provide more space to family members, friends and guests during the extravagant aristocratic social gatherings.994 It is also been suggested that the wealthy domini at times may have opened their private baths to local community members in return for a

987 BOWES, K., 2006, p. 83. 988 , J.; ETIENNE, R. ; MAYET, F. ; et alii, (Portugal), Paris, 1990. 989 GORGES, J.G., - , Paris 1979. FERNÁNDEZ-GALIANO, D., “La villa romana de Carranque,” in Hispania. El legado de Roma. Madrid 1999, pp.487-489; Idem (ed.), Carranque. Centro de Hispania Romana. Alcalá de Henares, 2001. See also CARRANQUE 2001. 990 GORGES 1979. 991 For a discussion on villas and cult see BOWES, K., Possessing the holy: Private churches and private piety in late antiquity, PhD Thesis, Princeton University, 2001; idem “Building Sacred Landscapes: Villas and Cult”, in Villas Tardoantiguas en el Mediteráneo Occidental, (ed. CHAVARRIA A.-ARCE J.-BROGIOLO G.P.), Madrid 2006, pp. 73-95. See also: BOWES, K., Private Worship, Public Values and Religious Change in Late Antiquity, 2008. 992 BOWES 2001, p. 95. 993 For paganism at this time see Mac MULLEN 1981 and 1997; LEE 1999. 994 For an analysis of the double set of baths at Palazzi di Casignana see Chapter 4 of this study. 267

fee, especially during the latest Roman centuries when towns were no longer able to provide public baths, or to renovated the existing ones.995 On the other hand, the construction of double baths in Roman villas may have also been determined by a religious motivation and diffusion of Christian ideology. Literary sources indicate a preference towards private baths where, contrary to what occurred in public baths, it was possible to preserve one’s modesty.996 A reinforced notion of body care and social relationships may have sparked the need for more private experiences, hampering the communal character of Roman society.997 More modesty may have driven the need for larger private thermal facilities, duplicating them within the same residence, or increasing the number of tubs to allow individual immersions preserving the modesty of the bather, especially inside luxurious villas where the availability of extensive areas made it possible.998 It is important, when analysing late Roman architecture, to keep in mind that the Roman upper classes had changed. While the elite were still acquiring and nourishing classical culture, they were also absorbing Christian ideology, which was being expressed in the lavish residences and power-architecture of the Late Empire.999 B. Lancon reports the Christian conversion of the urban prefect Furius Maecius Gracchus, a descendent of the Gracchi. When he was appointed prefect of the city in 376, according to Jerome and Prudentius, during his term of office he had himself baptised as well as his lictors. In addition, he ordered the destruction of pagan cult images and the closing of one of the several sanctuaries of Mithras.1000 This impressive conversion of a prestigious personage such as Furius Maecius Gracchus indicates the progress of Christianity among the elite of Rome. Lancon points out how in the time of Valentinian II (371- 392), Ambrose said that the Christians had reached a majority in the .1001 The same is found in Prudentius around 400. Prudentius in his poem against Q. A. Symmachus writes:

“There are hundreds of houses of old noble blood, who have been marked with the seal of Christ, thus escaping the vast abyss of shameful idolatry.” And a little farther on: “Rome serves the God-Christ and rejects its old cults.”1002

The following sections will look at the literary and archaeological evidence of private chapels in Roman domestic architecture. Private Christian worship was becoming a widespread phenomenon in the fourth century AD, and this may have involved the owner of Villa at Palazzi di Casignana.

995 LIPPOLIS 2001, pp.64 ff. 996 Carmen 23, pp. 495-499. 997 GINOUVES 1955; CAGIANO DE AZEVEDO 1972, p. 357. On the conception of body and self-care that pre- dates Christianity, see FOUCAULT 1986; CAMERON 1986; BROWN 1988. 998 Private baths found in urban contexts were not as large. See houses in Aphrodisia, Antioch, , Athens, Bulla Regia and Carthage. BALDINI LIPPOLIS 2001, p.65 note 203-208. On the other hand, in the country of many regions of the Empire, the tendency was to increase the size of the baths and create complex plans. For Hispania, see GARCIA-ENTERO, V., “Los Balnea de las Villae Tardoantiguas en Hispania,” in CHAVARRIA- ARCE-BROGIOLO 2006, pp. 97-111; For Aquitania, see BALMELLE 2001, pp.187-196.. 999 SALZMAN, M.R., The Making of a Christian Aristocracy: Social and Religious Change in the Western Roman Empire, 2002. 1000 LANCON 2000, pp. 71 ff.; Jerome, Letters, 107, 2; Prudentius, Against Symmachus, I, 561 ff. 1001 LANCON 2000, p.72; Ambrose, Epistulae, 17.10. 1002 Prudentius, Against Symmachus, I, 566-568, and II, 440-441. LANCON 2000, p. 72. 268

Literary evidence for private chapels in late Roman houses

After the Peace of the Church (312), and even more during the years when Christianity became the Empire’s official, solely sanctioned cult under Theodosius (379-395 AD), Christians seem to have been motivated to build religious spaces inside their homes for private worship, a phenomenon that started with the rise of Christianity. This custom is inserted in a long tradition of religious facilities associated with houses. Pagan shrines were built inside imperial palaces, villas, and houses and came in different variants according to specific needs and historical- geographical contexts. They were found at the Diocletian Palace of Split;1003 the Palace in Nicomedia,1004 the complex of Maxentius on the Way,1005 and inside the villa at Montmaurin.1006 For many converted members of the Roman aristocracy who could afford the expense, to create spaces devoted to Christianity in their houses must have been a natural choice. In several instances ancient texts record that wealthy late Romans built inside their residences chapels for private worship. These testimonies supplement the shadowy archaeological evidence on private cult in the domestic sphere. The life of Melania the Younger, for instance, alludes to the existence of a private oratory inside her Roman house on the Caelian Hill, where she prayed the whole night prior giving birth to her second child.1007 Other texts report that members of the Spanish and Gallic aristocracy were eager to obtain relics of local and Holy Land provenances in order to place them in their properties, which contained sacred spaces built for that purpose (martyria).1008 Melania the Elder (ca. 341 AD –410 AD), grandmother of Melania the Younger, Paulinus of Nola and Suplicius Severus (ca. 363 AD – ca. 425 AD) owned a number of relics in their homes.1009 Paulinus of Nola (ca. 354 AD – 431 AD) described a church-mausoleum on his ancestral estate near Bordeaux attesting to a trend inside aristocratic properties that blended Christian cult and commemoration of the dead.1010 Sulpicius Severus tried to acquire the body of Martin of Tours but had to be content with a relic of Martin’s disciple Clarus, which he kept in a private church on his estate together with other relics from the Holy Land.1011 Owning Holy Land relics must have increased the status of his estate and monastery community.

1003WILKES, J., Diocletian's palace, Split: residence of a retired Roman emperor, 1986; MARASOVIC- MARASOVIC 1994. 1004 BOULHOL, P., L’Apport de l’hagiograhie à la connaisance de Nicomédie palechrétienne (toponymie et monuments) in MEFRA, 106, 1994, pp. 991-992. 1005 PISANI –SARTORIO- CALZA, La villa di Massenzio sulla via Appia: il palazzo, le opere d’arte, Roma 1976. 1006 BALMELLE 2001, p. 187; FOUET, G. et Alii, The Roman villa of Montmaurin: historic building, Saint Gaudence-France, 1986. 1007 GERONTIUS, Life of Melania, 4-6. 1008 BOWES, K. 2006, p.95; idem 2001; KULIKOWSKI, M., Late Roman Spain and Its Cities, Baltimore/London 2004. 1009 BOWES 2006, p.95; idem 2001. For the traffic of relics see also HUNT, E.D., “The traffic in relics: some late Roman evidence”, in HACKEL,S., The Byzantine Saint, 1981, pp.171-180. 1010 Paolinus of Nola, Epistulae, 12.12. 1011 BROWN, P., The cult of the saints, 1981, p.31. 269

Archaeological evidence: Private chapels inside urban residences

In contrast to the clear literary testimonies of private cult in the domestic sphere, archaeological evidence for Christian use and adaptation of private space for communal purposes from the third century onward is often blurry. The excavated material is not always comprehensive enough to determine whether a space inside a private residence was used for religious purposes or not. Yet, it is well known that Christians met in homes even prior to the fourth century and that Christianity was from the beginning a household movement. Christian groups relied upon the patronage and support of private patrons for places of worship and for money to support their projects. As Christianity began attracting members of the upper classes in the second century AD, these patrons were increasingly wealthy members who played an important role in the hierarchy of the community.1012 They provided a religious space for the community in the dining and reception rooms of their homes where they performed the rituals of the Eucharist and baptism. As a consequence the bond between these patrons and their community was greatly strengthened. A well dated and clear example of an urban residence adapted and dedicated to cultic use is provided by a house at the Syrian town of Dura-Europos, a frontier post on the Euphrates. The house’s owner devoted the entire ground floor to a communal meeting room for Christian religious purposes (fig. 10.2). Dining room G was enlarged to facilitate assembly and liturgical functions. Room D-E was transformed into a baptistery and its walls were painted with Christian frescoes (fig.10.3).1013 They are probably the most ancient Christian paintings, representing the ‘Good Shepherd’ (this iconography had a very long history in the Classical world), the ‘Healing of the paralytic’ and ‘Christ and Peter who walk on the water.’ A much larger fresco depicts two women approaching a large sarcophagus, i.e. probably the two Marys visiting Christ's tomb.

1012 BOWES 2008, p.49. 1013 MAIER, H.O., “Heresy, Households, and Diversity”, in A People’s History of Christianity, in BURRUS, V. (eds.), 2005, pp. 224-225. 270

10.2 Dura Europos, Syria: The third century AD house-church.1014

1014After http://www.pitt.edu/~tokerism/0040/syl/christian.html 271

10.3. Dura Europos, Syria: The baptistery of the third century AD house-church with the water tub for baptismal immersion.1015

From their origins, Christian groups had a dispersed hierarchy. Patrons, prophets, apostles, teachers, presbyters and bishops appear in second century sources as persons with various kinds of duties and authorities.1016 As Bowes points out, a Christian community within each territory looked at these figures as their spiritual leaders, a role that was exclusively claimed by the bishops after the Peace of the Church (312 AD). The origin of the episcopal office may lie in the need that a given community had to communicate with and be represented in front of other Christian centres of the Empire. Bishops being the descendants of the apostles had full control over communal properties and liturgies, especially the Eucharist and baptism. Yet teachers and charismatic spiritual leaders continued to rival bishops for community leadership. Among them were those who professed heresies, departing from Christian orthodoxy. In turn, bishops increased their supervision of private worship, claiming their exclusive authority over rituals and sacred gatherings. Constantine and his successors sanctioned the confiscation of the property of those patrons who gathered in their homes meetings of heretics. These laws underline how widely diffused was the phenomenon of early Christian communities worshipping in private spaces in conjunction with or outside the boundaries of official imperial Christianity. Bowes notes the absence of Christian basilicas within the boundaries of Rome’s walls and in the trans- Tiber neighbourhoods during the fourth and early fifth century AD, which indicates that Christians continued to attend weekly services in private domestic spaces as their ancestors did in earlier times. The great majority of the Constantinian basilicas were martyr-shrines and

1015 After http://www.pitt.edu/~tokerism/0040/syl/christian.html 1016 BOWES 2008, p. 49, note 200. 272 cemetery-churches built outside the city walls. This was true of even the great Lateran cathedral laid against the city walls. Christian architecture, even after the Peace of the Church, grew slowly and was modest in scale. It was only in the early fifth century that monumental, three- aisled basilicas appear with frequency.1017 Private, domestic Christian worship has been also suggested for other cities of the Empire where there is a lack of churches. F. Guidobaldi has underlined the high number of third and fourth century apsidal reception halls of aristocratic houses upon which churches were built centuries later. Apsidal halls at the churches of San Clemente, Santa Susanna, Santa Balbina and Quattro Coronati originally belonged to fourth century aristocratic houses. 1018 Guidobaldi argues that these apsidal halls were likely places for small-scale Christian meetings already in the fourth century AD, and that they gradually were replaced by proper churches. This assumption is based upon architectural features, despite the total absence of fourth century liturgical apparatus or Christian decoration inside these late Roman residences on top of which churches were later built. A social factor suggests that apsidal halls in several cases were spaces associated with cultic use. In fact, aristocratic homes of Rome were the centre of daily Christian life even a century after the Peace of the Church (312 AD), until public churches were built in early fifth century. As Bowes points out, Constantine did not leave the Church with an excess of cash for building churches, and Rome’s wealthy upper classes converted to Christianity more frequently towards the end of the fourth century. Thus, they began to contribute to the construction of public churches later. Even when the bishop of Rome and the clergy became more firmly established after the Pre-Nicene Peace (312 AD), Christians continued to gather in homes of wealthy patrons for ritual prayers, and the liturgy of baptism and Eucharist.1019

1017 BOWES 2008, p.72, note 81. 1018 GUIDOBALDI, F., “Roma. Il tessuto abitativo, le ‘domus’ e i ‘tituli,’ in Storia di Roma III. L’etá tardoantica. 2. I luoghi e le culture, in Carandini- Cracco Ruggini-Giardina (eds.), Turin 1993, pp. 69-83. 1019 BOWES 2008, pp. 71-75. 273

10.4. Palace of the Dux in Apollonia, modern Susa in Libya (early 6th c.).1020

Domestic chapels were found also at the Palace of the Dux in the Libyan city of Apollonia (modern Susa), and at the Palace of Ephesus in Turkey. The chapel at the Palace of the Dux (fig.10.4) is dated to the early sixth century and measures 9 m. x 7 m. It was placed in a remote side of the house along the peristyle and featured three aisles, a large narthex and an eastern apse flanked by two side rooms. A reliquary found inside this room has also suggested the use of this space as martyrion.1021 The chapel inside the Palace of Ephesus was set in a distant corner of the house. It was smaller than the chapel of Apollonia and had a single aisle. The apse contained a synthronon, a stepped bench for the clergy. 1022 The small size and remote location of these two chapels has suggested a private use for domestic worship.1023

Private chapels inside suburban villas Private congregations and rituals inside private houses were common from the beginning of Christianity and developed in conjunction with the public religious ceremonies. The ecclesiastic organization in the countryside started in the fourth century and developed in the fifth century AD in most areas of the Empire.1024 If ritual spaces were built while the villa was

1020 After BURRUS, V. (ed.) 2005, p.201, fig.83. 1021 GOODCHILD, R., “The Palace of the Dux,” in HUMPHREY, J. (eds.), Apollonia: The Port of Cyrene: Excavations Conducted by the University of Michigan 1965-1967, Supplements to Libya Antiqua 4 (Tripoli:Derectorate General of Antiquities, Museums, and Archives, 1976), pp. 245-256; ELLIS, S., “The Palace of the Dux at Apollonia and Related Houses,” in BARKER, G., LOYD, J., and REYNOLDS, J., (eds.), Cyrenaica in Antiquity, Biblical Archaeology Review International Series 236 (Oxford: Biblical Archaeology Review, 1985), pp. 15-25. 1022 BOWES 2005, pp. 204-205. 1023 BOWES 2005, pp.200 ff. 1024 BROGIOLO-CHAVARRIA ARNAU 2005, p.127 ff. 274 still functioning, it is likely that the dominus was the person in charge of the chapel construction and perhaps the cult itself.1025 As Bowes points out, Tertullian (On Prayer 19.2-4), Hyppolitus (Ap. Trad. 36), Novatian (On Spetacles 5.4-5; Jerome, Ep. 48.15) and others attest to the practice of private prayer and the reading of the sacred texts in private houses. Even the consumption of a portion of the Eucharist, typically the bread only, took place in private homes. The worshippers would then bring the Eucharist home from the Sunday mass to eat it during the week.1026 Late antique sources make it clear that by the late fifth century private chapels and churches were run by clergy who were nominated by the villas’ patrons and chosen among the estate tenants and working dependents, but officially ordained by the episcopate. Patrons had the power to appoint, sustain and exercise authority over the clergy, in the same manner as they would over other dependants. Private worship led by local aristocracies was so widespread that from the fifth century onward conflicts arose between these wealthy patrons and the ecclesiastic authorities. The Church eventually took full control over the Christianization of the rural areas by establishing and supervising churches and baptisteries.1027 Church councils in Gallia during the fifth and sixth century prohibited villa owners from celebrating Eucharistic masses in their estate-churches on Christmas, Easter and Pentecost and other major festivities, insisting that they go instead to the bishops in the main cathedral churches.1028 This decree indicates that villa-churches held religious gatherings which were attended by the villa owner’s family and the estate’s community at large. The phenomenon of worship within the domestic sphere must have spread uncontrollably if the episcopate at some point was compelled to enforce restrictions.1029 Since the territory of Locri was fully Christianized in the fourth century AD, it is possible that the Villa at Palazzi di Casignana was the seat of religious gatherings under the control of the villa’s dominus. The construction of a religious space inside the villa might have given prestige to the rural residences of those aristocrats that held public positions.1030 By the later fourth century aristocrats such as the Gallic statesman Ausonius built private chapels not only in their urban residences, but also in their country villas. Ausonius tells us of his habit of opening an undecorated chapel in his villa of Aquitania, which he calls sacrarium, before going out.1031 The Goth general Flavius Valila built a church for the community of that territory, the ecclesia Cornutatensis, in his property, the massa Cornutiana, near his praetorium in Tivoli.1032 The Codex Theodosianus and the Canons of the Elvira Council refer to the responsibility of the landowners to take care of the faith of peasants who lived in their estates. It can be assumed that they often provided spaces for worship.1033

1025 CATINO-WATAGHIN 2000. 1026 BOWES 2005, p.207. 1027 BROGIOLO-CHAVARRIA 2005, p.128; PERGOLA 1999. 1028 BOWES 2005, p. 207. 1029 The domestic sphere may also have encouraged the spread of heretical ideas. The Codex Thedosianus in several edicts shows the concern for heretical gatherings in private houses establishing as form of punishment the confiscation of the properties (MAIER 2005, pp. 224-225; BOWES 2005, p.207). Both scholars refer to CTh. 16.5.3 [372]. 16.5.9.1. 1030 BROGIOLO-CHAVARRIA 2003, p. 29. 1031 For a description of the palaces See BALDINI LIPPOLIS 2001, p.66 with bibliographical notes. Ausonius Efem, II.7: …pateatque fac sacrarium nullo paratu extinsecus…”; 1032 VERA 1999, p.1019 and note 112; Liber Pontificalis 116. 1033 Cod. Theod. 16.5.14; Elvira canons 40,41, 49; see SFAMENI 2006, p. 270. 275

Yet only a few private chapels have been identified with a high degree of certainty in villas, as we have seen in the previous chapter. In Spain, the Villa of Fortunatus in Huesca, which we have already mentioned (fig. 9.30), features a late fourth century chapel that was built over an earlier three-aisled triclinium.1034 In the same province, another private chapel was individuated inside the Villa Marialba outside Leon.1035 Other suspected private chapels in Spain appear to be the cross-planned structures found, for instance, inside the Villa of Torre Águila (Badajoz) dated between the fourth and fifth century which has not yielded evidence of Christian liturgical use,1036 or inside the late fifth century villa of Monte da Cegonha.1037 In Britain, the Villa of Lullingstone in Kent (fig.9.34) dated to 400 AD, features a Christian chapel in a distant corner of the villa, slightly off the main circulation axes. This was built over a cellar that was previously occupied by a pagan shrine. The room was rectangular and fairly small measuring only 7 m. x 4 m. It featured an eastern niche (perhaps for holding a cultic statue), and was surrounded by a series of small rooms including an antechamber and side chambers.

10.5. Kent: Villa at Lullingstone (ca. 380 AD). Painting with a chrismon.1038

1034 PALOL, P. “La arqueología cristiana en la Hispania romana y visigoda. Descubrimientos recientes y nuevos puntos de vista,” Actes du XI congres international d’archeologie chretienne,vol.2, Lyon, 1989, 2001-2003. 1035 BOWES 2001. 1036 See MARTIN, R., “La villa romana de Torre Águila a partir del siglo IV d.C. Considerationes generales,” Congreso Internacional. La Hispania de Teodosio, Segovia, 1997, pp. 607-711. 1037 ALFENIM, R. - LOPES, C., “A basílica paleocristã/Visigótica do Monte Cegonha (Vidigueira),” IV Reunión d’Arquelologia Cristiana Hispánica, Lisbon, 1994, pp.389-399. 1038 After http://www.religionfacts.com/christianity/symbols/chi_rho.htm. A description of this villa is in BURRUS, V. (ed.), 2005, p.203, fig.8.5. 276

The walls were painted with unidentified scenes that included three large representations of the Chi-Rho, or chrismon, which clearly symbolize Christ (fig.10.5). Their placements mark the processional way of the worshippers from the entrance to the chapel.1039 A number of other candidates in other provinces of the Empire have been proposed as examples of later fourth and early fifth century villa-chapels, largely without proof of Christian liturgical function.1040In Italy, cultic spaces inside late Roman villas have been recognized in a handful of cases. At the villa in Desenzano, the large apsidal reception room paved in opus sectile (fig.5.19, Room 35) dated between the fourth and fifth century AD, contained an infant burial in the shape of an elongated pentagon.1041 The location slightly off the centre from the apse and the care with which the tomb was constructed led the excavators to suggest that this room between late fourth century and the fifth centuries was used for worship as a private oratory.1042 Other findings would confirm that the dominus of Villa at Desenzano was converted to Christianity: several lamps decorated with Christian motifs, a fragment of glass engraved with the figure of Christ and the rooster; a mosaic with a scene representing a young man surrounded by several animals, an image that has been interpreted as the Good Shepherd.1043 A private oratory inserted in a villa that was certainly still functioning was found in Sizzano, near Novara (Piemonte),1044 and another was discovered inside a late Roman villa of Abruzzo in central Italy.1045 A text written by Vigilius, the first bishop of Trent, attests to the construction of a cult edifice in the Late Roman countryside of northern Italy, which has not been yet confirmed by archaeological findings.1046 Vigilius wrote that his companions Sisinnius, Martyrius and Alexander founded an ecclesia in Val di Non at the end of the fourth century AD. In this building worshippers read and commented the sacred texts, prayed together but could not celebrate the Eucharist because they did not have a presbyter (in the New Testament this word

1039 MEATES, G.W., The Roman Villa at Lullingstone, Kent, vol.1: The Site, and vol.2: The Wall Paintings and Finds, Monograph Series of Kent Archaeological Society 1 and 3, 1987. 1040 M. Bassani has recently studied domestic spaces devoted to cultic use in Spain. BASSANI, M., “Ambienti e edifici di culto domestici nella Penisola Iberica, Pyrenae, 36.1, pp.71-116. 1041 SCAGLIARINI CORLAITA et Alii, Villa romana di Desenzano. Itineari dei musei, gallerie, scavi e monumenti d’Italia, 1992; LA GUARDIA 1994. 1042 MIRABELLA ROBERTI 1994, p.110-111. D. Scagliarini Corlaita disagrees with this interpretation by pointing out that it is more frequently attested to find a villa or part of it transformed into a church rather than a villa and chapel functioning at the same time. SCAGLIARINI CORLAITA 1994, p.55. A different view is expressed by RIPOLL-ARCE 2000. See also SFAMENI 2006, p.257. 1043 SANNAZZARO, M., La cristianizzazione delle aree rurali della Lombardia (IV-VI sec.). Testimonianze scritte e materiali, 1990, pp.32-33. 1044 SFAMENI 2006, pp. 257-258; CANTINO WATAGHIN - UGGÉ 2001, p.23. 1045 SFAMENI 2006, pp. 257-258; COLECCHIA 2000, p.107. 1046 For the Christianization of rural northern Italy see CANTINO WATAGHIN 2000. For the following discussion See BROGIOLO-CHAVARRIA ARNAU 2005, p.127 ff. According to tradition, Vigilius of Trent he was a Roman patrician educated at Athens. Perhaps he was friend of Saint John Chrysostom. He went to Rome and afterwards settled at Trento in 380 and was chosen to be bishop of that city. He may have been consecrated by either Ambrose of Milan or Valerian of Aquileia (Valerianus). As bishop, Vigilius attempted to convert Arians and pagans to orthodox Christianity and is said to have founded thirty parishes in his diocese. A letter attributed to Ambrose tells that Ambrose encouraged Vigilius to oppose marriages between Christians and pagans (Ep. 29 in P.L., XVI, 982).Vigilius preached in Brescia and Verona, which lay outside of his diocese. His companions during his missions were Saints Sisinnius, Martyrius and Alexander, who were sent by Ambrose to assist Vigilius. Tradition makes these three natives of Cappadocia. A work called De Martyrio SS. Sisinnii, Martyrii et Alexandri (P.L., XIII, 549) is attributed to Vigilius. 277

refers to a leader in local Christian congregations). M. Sannazzaro has argued that the construction of a church in a remote area of Northern Italy such as the one in Val di Non, indicates that very likely other churches were being built in the countryside during four century AD even if the archaeological documentation on this type of buildings is almost nil.1047 In northern Italy, another cultic space has been identified inside a late fourth century villa in Palazzo Pignano near the town of Cremona (fig. 10.6).

10.6. Cremona, northern Italy: Palazzo Pignano (5th c.).1048

1047 BROGIOLO, CANTINO WATAGHIN, GELICHI 1999, pp. 533-534. 1048 After MASSARI,G. – ROFFIA, E., La villa tardo romana di Palazzo Pignano (Cremona), in G. PONTIROLI (ed.), Cremona Romana, Atti del congresso storico archeologico per il 2200 anno di fondazione di Cremona (Cremona 1982), Cremona 1995, pp.185-227. 278

A series of apsidal, circular and squared rooms is distributed along the octagonal central peristyle of the complex (fig. 10.6). On the eastern side, there is a set of structures that include a rectangular room, followed by smaller rooms and an apsidal hall. These rooms were all heated by an hypocaust system.

10.7. Cremona: the Romanesque church built on top of a late 4th c. circular room that was part of a Roman villa.1049

On the western side of the complex there is a Romanesque church (fig.10.7). This was built on top of a circular room that belonged to the late fourth century Roman complex. (fig.10.6: see bottom of the picture). One of the rooms that flanked the circular room contained a tub that was identified as a baptismal font.1050 Thus, G. Cantino Wataghin maintains that the round room was a private church built inside the residential villa by a late fourth century owner.1051 The name ‘Pignano’ has brought scholars to hypothesize that the owner of the late Roman edifice was Pinianus, first cousin and husband of Melania the Younger. He belonged to the late Roman senatorial family of the Valerii. This hypothesis would be supported by another coincidence. The oldest church in Rome that features a round plan, the Church of Santo Stefano, was built in the fifth century AD on the Caelian Hill, where a huge villa owned by the Valerii was found.1052 According to Mirabella Roberti, Palazzo Pignano may have been the rural seat of the local bishop as even the modern name ‘Palazzo’ (Palatium) seems to suggest.1053

1049 After http://it.wikipedia.org/wiki/File:Palazzo_Pignano.jpg. 1050 PASSI PITCHER 1990 a; idem 1990 b; SFAMENI 2006, pp. 243-247. 1051 CANTINO WATAGHIN 1994, p. 146. 1052 MASSARI, ROFFIA 1995; PASSI PITCHER 2002. A study on the Christianization of the domus of the Valerii on the Caelian Hill is in BRENK 1999. See also BRENK 2003. 1053 MIRABELLA ROBERTI 1971, p. 344. 279

A. D. Lee has shown that much of the surviving epigraphic evidence about bishops reflects their role as builders, whether of churches or of other structures for the wider community. Constantine bestowed on them extra ecclesiastical power, which made them prominent figures within their communities until their death. Moreover, bishops were drawn from the local elites who had been used to providing community leaders for centuries.1054 As Lee points out, by the 250s there were adherents of Christianity among the Roman social elite – senators and equestrians. Several inscriptions from Rome and a number of Christian writers such as Tertullian indicate how Christianity had penetrated the administrative ranks of the imperial palace in the early third century. Archaeological discoveries of buildings such as Palazzo Pignano further attest the Christianization of the late Roman elite and how this elite made an effort to create religious spaces, private oratories and chapels inside their rural residences.

1054 LEE, A.D., Pagans an Christians in Late Antiquity. A Sourcebook, 2000, pp. 213 ff. 280

Conclusions

In the fourth century AD, Christianity was rapidly developing within the urban Episcopal centres and in the rural domain of rich landowners, including territories such as that of Villa at Palazzi di Casignana. So far there is not sufficient data for us to assert that the owner of the villa was Christian. Yet, the lack of evidence for Christian activity inside the villa may be due to a limited archaeological investigation on the site, or to the loss of archaeological findings, rather than to a factual absence of this material. The villa was first discovered in 1964, but a solid scientific method of excavation on the villa’s site has been employed only in the most recent archaeological campaigns. Furthermore, much destruction occurred in the past when the modern road S.S. 106 and the train track were built. Extensive plunder at the site may have also occurred over the centuries. A huge amount of material evidence has been forever lost making it difficult to reconstruct the history of the villa’s estate. The hypothesis addressed in this study that Room I was a chapel, may encourage future researchers to focus on graves placed inside and around the villa and also to further explore the function of Room D and E, which were transformed to contain water, and the third century structures that lay underneath Room I, including the octagonal Room V.

281

CONCLUSIONS

The grandeur, opulence and sumptuousness of the Villa at Palazzi di Casignana is evidence that refutes the belief that Locri and the surrounding area during the late imperial period suffered severe economic decline and impoverishment. The presence of this remarkable villa forces us to revaluate the long-standing preconception that the southern Italian region of Italy was largely insignificant and unworthy of scholarly attention during the late imperial period. This study explains how the abundance of natural resources in the region prompted the late Roman elite to invest in and exploit this region.1055 The villas near Palazzi di Casignana, in the territory of Locri provide further proof that during the late imperial time Bruttii was led by a lively and active elite class, who exploited the natural resources in the region being favored by an articulated road system and access to important Roman harbors. Besides the enormous scale, the wealth of Villa at Palazzi di Casignana is exemplified by more than 20 elegant floor mosaics, glittering vault mosaics that likely decorated the ceilings, and by lavish and exquisite marble veneers that covered the walls of several rooms. This luxurious decorative apparatus and the iconographic themes of the mosaics stand as the most evident proof that the fourth century villa’s owner shared common artistic taste and sophisticated paideia with his peers in the rest of the Western Empire. This is not a lifestyle described by the traditional scholarship that depicted Bruttii as impoverished. There are several buildings near the Villa at Palazzi di Casignana, suggesting that the entire site was economically productive during the inhabitation of the villa from date to date. The wealth became even more robust during the fourth century AD when the latest monumental architectural renovation took place. This study acknowledges that the villa could have been a statio. Archaeological evidences and ancient sources however support the hypothesis that the villa was instead a private residence of a fourth century dominus who might have had strong economic and political interests in the area of Locri, perhaps in the rest of Bruttii and with Rome. The name “Palazzi di Casignana,” of the modern municipality to which this villa belongs, may have derived by ‘palatium,’ namely a luxurious residence, and the name ‘Casignana’ may come from ‘Casinius’ a Roman landlord.1056 The enormous size and monumentality of the villa are features not necessarily associated with a statio. Late Roman sources and several archaeological discoveries --- for instance, the late Roman villas at Piazza Armerina, Patti and Tellaro in Sicily, the villa at Desenzano in Northern Italy, the one at Faragola in Puglia and those in Spain and France, show that Roman villas across the Empire expanded dramatically to reach the size of towns in the fourth and fifth centuries AD. By combining otium (leisure) with negotium (business) these villas served not only for retreat but also likely operated as self-sufficient farming and trade enterprises. The buildings uncovered near the Villa at Palazzi di Casignana very likely included dwellings of peasants, facilities for the production and storage of agricultural goods as wine, olive oil, fruit, or stables for cattle rearing. They may attest that the villa’s owner was likely involved in farming and trade. Because of the large amount of wine amphorae Keay 52 discovered at the villa’s site, which have been produced in Bruttii and exported all over the

1055 Based on ancient texts such as the Expositio totius mundi, wine was likely one of the most important sources of revenue in Bruttii. 1056 This hypothesis has been proposed by BARILLARO 1979. 282

Empire, it is highly possible that the dominus engaged in wine production during the fourth century AD. Another economic activity that may have occurred at the site is fish-rearing, if archaeologists can prove that the water tanks located behind the monumental nymphaeum were used for this function. The location of the villa near important roads, the River Bonamico and the Ionian Sea likely facilitated the agricultural and industrial manufactures on the villa’s estate. Deeper explorations of the facilities near the villa could bring important distinctions to light, and provide evidence that Palazzi di Casignana was a vital heart of the fourth century economy in the Locri’s area and in Bruttii. In Late Antiquity, as for earlier times, several prestigious Roman families lived in Bruttii and in Locri’s area. Under Diocletian, Bruttii and Lucania formed one of the hundreds of provinces that this emperor instituted in order to facilitate the administration of the Empire. Rhegium was the main capital city of Bruttii and was chosen as the seat of the corrector (governor) of Lucania et Bruttii. Rhegium was an extremely important port of the Mediterranean that opened Bruttii to the trade and commerce with many distant regions of the Empire and especially with Africa, which was an economic driving force during the late imperial period. Eminent figures such as the senator and consul Q. A. Symmachus (4th. c. AD) and Cassiodorus (6th c. AD) were appointed to the magistracy of the governorship of Lucaniae et Bruttii. Symmachus was corrector in 365 AD while his father, L. Aurelius Avianius Symmachus, was prefect of Rome.1057 He likely held this governorship during the same time when the Villa at Palazzi di Casignana reached its splendor. Symmachus’ letters tell us that he owned estates in many areas of the Empire, and very likely even in Bruttii, while he was corrector at Rhegium in 365 AD. Cassiodorus, who belonged to the same senatorial family of the Aurelii, was corrector of the province from 515 to 523 AD. He is known to have owned estates in Bruttii and in his native Ionian town of Scolacium where he founded the Vivariensis monastery.1058 The presence of these two prestigious senatorial aristocrats (and of the correctores who came before and after them), in Brutiii and in Rhegium must have created a social milieu and aristocratic entourage in Bruttii throughout Late Antiquity. This, in turn, must have had an impact on the economy and land occupation across the region, especially in the towns near Rhegium, such as Locri and Gioiosa Jonica. It is possible that the Villa at Palazzi di Casignana may have belonged to a late Roman wealthy aristocrat who had some association with prominent senatorial exponents such as Symmachus. Their political role in Bruttii is highly relevant. These governors typically conciliated powerful local families while holding their office and developed their own social network that extended across the Empire. It has been argued that there was a correlation between the importance of a province and the political and social stature of the magistrate appointed to its governorship.1059 This would confirm that throughout Late Antiquity Bruttii was economically vital and provided favorable political and economic opportunities to the Roman senatorial aristocracy and to the elite across the Empire. As they invested their fortune in this southern region, they likely increased their wealth and strengthened their political power. Another important aspect is the possible relationship between the Villa at Palazzi di Casignana and the spread of Christianity in the territory of Locri between the third and fourth

1057 MATTHEWS, J., Western Aristocracies, 1990, p.14. 1058 ZINZI 1998. 1059 CECCONI, G.A., Governo imperiale e èlites dirigenti nell’Italia tardo antica: problemi di storia politico- amministrativa, Como 1994. 283

century AD. Locri became a Diocesis in 330 AD. This fact alone proves that Christianity was already institutionalized and pervasive in the territory. The proposal that the apsidal and cruciform Room I inside the Villa at Palazzi di Casignana may have been a chapel is then legitimate and worthy of deeper exploration. During Late Antiquity, Roman aristocrats such as Ausonius, Sidonius and Melania the Younger, built private spaces inside their residence for Christian worship and some may have even opened these religious spaces to the community on their estate. This may be the case for the owner of the Villa at Palazzi di Casignana. He may have built the monumental and cruciform Room I to host religious gatherings inside his private residence, a practice that was not uncommon at this time. Even the construction of two separate baths, the Western and Eastern Baths of the villa, which is a feature encountered in other well known fourth century villas of the Empire, may have been strongly influenced by Christianity, which banned the mixed bathing and promoted modesty among the new converted. The presence of burials around the villa would offer another valid argument that the Villa at Palazzi di Casignana may have hosted a Christian cult during the fourth century AD. It is possible tha the dead buried near the Villa at Palazzi di Casignana were in large part Christian because when the villa was renovated in the fourth century Locri and the rest of Bruttii was largely Christianized. The burials may gave belonged to a community that gravitated around the villa during the late imperial period. Communal cemeteries typically house deceased that share a collective identity. The people buried near the villa might have shared the same religious beliefs and identified themselves with the villa and its estate perhaps even more so because the villa had religious significance to their lives. According to Yasin, literary sources indicate that by the third century AD there were cemeteries dedicated exclusively to Christian burial, such as the areae of Rome and Carthage.1060 In 203 AD Tertullian talks about “areas of our burials” (Ad Scapulam, 3.1) indicating that Christians at his time were buriying their dead in their own necropolises. Even Origen mentions the existence of Christian cemeteries in already at the beginning of the third century AD (Homily on Jeremiah, 4.3.16).1061 Much more exploration is required to validate all of these hypotheses. It is to be hoped that the Villa at Palazzi di Casignana will inspire a fresh and more fruitful interest towards the Roman history of Calabria among contemporary scholars and people in general. Important sites in Calabria, such as the Villa of Naniglio, the Villa at Marina di Gioiosa Jonica and in Acconia di Curinga deserve to be revaluated, further studied, published and, ultimately, elevated from their state of abandonment and neglect.

1060 YASIN 2005. 1061 Ibidem, with a thorough bibliography on the formation of Christian cemeteries. 284

BIBLIOGRAPHICAL REFERENCES FOR THE MAIN ANCIENT SOURCES.

The ancient sources are cited following the abbreviations of the Thesaurus Linguae Latinae (index) Lipsiae 1904; HG. Liddle-R. Scott, A Greek-Engglish Lexicon, Oxford 1961; and G.W.H. Lampe, A Patristic Greek Lexicon, Oxford 1961(Christian-Greeks).

St. Ambrose Epistulae: G. Banterle (ed.), Sant’Arnbrogio, Discorsi e Lettere II/I, II/II, II/III, (Città Nuova), Milano-Roma 1988;

Ammianus Marcellinus Historiae: G. Viansirio (ed.), Ammiano Marcellino, Storie, I-III, (Oscar Mondadori), Milano 2001-2002.

Ausonius De herediolo, Mosella, Epigrammata, Epistulae: H.G.E. White. (ed.), Ausonius, I-II (Loeb), London 1949.

Cassiodorus Institutiones: R.A.B. Mynors (ed.), Cassiodori Senatoris Institutiones, Oxford 1937. Variae: A.J. Fridh (ed.), Variarum Libri XII, (Corpus Christianorum, Series Latina 96), Turnholti 1973.

Cato De agricultura: L. Canali-E. Lelli (ed.), Catone ii Censore, L’agricoltura, (Oscar Mondadori), Milano 1990.

Codex Theodosianus P. Kruegger (ed.), Codex Theodosianus, Berolini 1923.

Columella De agricultura: E.S. Forster-E.H. Heffner (ed.), Columella, On Agricolture, 1-3, (Loeb), London-Cambridge MA, 1954-1958.

Ennodius Opera: F. Vogel (ed.), Monumenta Germaniae Historica 7, Auct. Antiq., Berolini 1961.

Eusebius Vita Constantini: L. Tartaglia (ed.), Eusebio di Cesarea, Sulla vita di Costantino, Napoli 1984.

Expositio totius mundi, J. Rougè (ed) Expositio totius mundi et gentiun, (Les editions du Cerf), Paris 1966.

285

Gregorius Magnus Epistulae: V. Recchia (ed.), Opere di Gregorio Magno, V/3, Lettere, (VIII-X), (Città Nuova) Roma 1998.Dialogi: B. Calati et alii (ed.), Opere di Gregorio Magno, IV, Dialoghi (I-IV), (Città Nuova), Roma 2000.

Gregory of Nissa Epistulae: P. Maraval (ed.), Grégoire de Nysse, Lettres, (Sources Chrétiennes 363), Paris 1990.

Gregory of Tours Historia Francorum: M. Oldoni (ed.), Gregorio di Tours, La storia dei Franchi,Milano 1981.

St. Jerôme Epistulae: J. Labourt (ed), Saint Jerôme, Lettres, vol. 1-8, (Les Belles Letters), Paris 1949-1963.

Justinian Digesta: A. Watson (ed), The Digest of .Justinian, Latin Text edited by Th. Mommsen, (Berlin 1868-1870), 1985.

Itinerarium Antonini Augusti 0. Cuntz (ed.), Itineraria romana. I. Itineraria Antonini Augusti et Burdigalense, Stutgardiae 1929.

Libanius Epistulae: R. Foerster (ed.), Libanii opera, vol. X, Epistulae 1-839, Lipsiae 1921. Orationes: J. Martin-P. Petit- P.-L. Malosse (ed.), Libanius, Discours, 1-4, (Les Belles Lettres), Paris 1979-2003. L. Harmand (ed.), Libanius, Discours sur les patronages, Paris 1955.

Liber Pontificalis (LP) L. Duchesne (ed.), Le Liber pontificalis, I-Ill, Paris 1955-57.

Olympiodorus of Thebes ap. Photius, Bibliotheca, cod. 80 Henyed = R. Henry (ed.), Photius, Bibliotèque, tome I, (Les Belles Lettres), Paris 1959, 166-187.

Palladius Opus agriculturae: R. Martin (ed.) Palladius, Traité d’agriculture, tome I (livres I ciii), (Les Belles Lettres), Paris 1976.

Paulinus of Nola Epistulae: G. Santariello (ed.), Paolino di Nola, Le Lettere, I-II, Marigliano 1992.

Paulino of Pella Eucharisticus: A. Marcone (ed.), Paolino di Pella, Discorso di ringraziamento, Fiesole 1995.

286

Pliny the Elder Historia naturalis: GB. Conte (ed.), G. Plinio Secondo, Storia naturale, I-V, Torino 1982-1988.

Pliny the Younger Epistulae: F. Trisoglio (ed.), Plinio Cecilio Secondo, Opere, I-II, Torino 1973.

Prudentius Contra orationem Symmachi: H.J. Thomson (ed.), Prudentius, II, (Loeb), London 1949-53.

Seneca Epistulae: F. Solinas (ed.), Seneca, Lettere morali a Lucilio, I-II (Oscar Mondadori), Milano 1995.

Sidonius Apollinaris Epistulae: W.B Anderson (ed), Sidonius, Poems and Letters, I-II, (Loeb), London 1956. Carmina: A, Loyen (ed.), Sidoine Apollinaire, tome II, Lettres (Livres I-IV,), tome III, Lettres (Livres VI-IX), (Les Belles Lettres), Paris 1970.

Sulpicius Severus Vita Martini: C. Mohrmann et alii (ed), Vita di Martino, Vita di Ilarione, In me- maria di Paola, (Fondazione Lorenzo Valla), Roma 1975.

Symmachus Epistulae: J.P. Callu (ed), Symmaque, Lettres, tome I (Livres I-li,), tome II (Livres IIl-V), (Les Belles Lettres), Paris 1972-82.

Vita Melaniae D. Gorce, (ed.), Vie de Sainte Mélanie, (Sources Chrétiennes 90), Paris 1962 (greek text). E. A. Clark, The Life of Melania the Younger: Introduction, Translation and Commentary, 1984.

Vitruvius De architectura: L. Bigotto (ed.), Marco Vitruvio Follione, Dc Architectura, libri X, Pordenone-Padova 1990.

287

BIBLIOGRAPHY OF SECONDARY SOURCES For the abbreviations of journals see the Année Philologique.

ACCARDO 2000 = S. Accardo, Villae Romanae nell’ager Bruttius. Il paesaggio rurale Calabrese durante ii dominio romano, (Studia archaeologica 107), Roma.

ACKERMAN 1992 = J.S. Ackerman, La villa. Forma e ideologia, Torino (ed. orig. London 1990).

ADAMS- R. LAURENCE= C. Adams and R. Laurence, Travel and Geography in the Roman Empire, (London ; New York : Routledge 2001),

ADAMESTEANU 1955 = D. Adamesteanu, Due problemi topografici del retroterra gelese, RAL s. VIII, 10, 199-210.

AGUILAR SÁENZ-GUICHARD 1993 = A. Aguilar Sáenz and P. Guichard, , La Sevillana et leur environnement, Madrid.

ALARCÃO-ÉTIENNE-MAYET 1990 = J. Alarcão, R. Étienne, F. Mayet, Les villas romaines de São Cucufate (Portugal), Paris.

ALFENIM-LOPES 1994 = R. Alfenim, C. Lopes, A basílica paleocristã/Visigótica do Monte Cegonha (Vidigueira), IV Reunión d’Arquelologia Cristiana Hispánica, Lisbon, 1994, 389-399.

ALLISON 2004= P.M. Allison, Pompeian Households: An Analysis of the Material Culture, University of California.

ANAMALI 1974 = s. Adhami, S. Anamali, Mosaïques de L’Albanie, Tirana.

ARCE 1986= J. Arce, El mosaico de “Las Metamorfosis de Carranque”, Madrider Mitteilungen 27, 365-374.

ARCE 1997a = J. Arce, Emperadores, palacios y “villae” (a proposito de la villa romana de Cercadilla, Cordóba), AntTard 5, 293-302.

ARCE 1997b = J. Arce, Otium et negotium: the Great Estates, 4th-7th Century, in L. Webster-M.P. Brown (ed.), The Transformation of the Roman World, AD 400-900, Berkeley, 19- 32.

ARNAUD 1993= P.Arnaud, “L’Itinéraire d’Antonin: un témoin de la littérature itinéraire du Bas-Empire,” Geographia Antiqua 2 (1993).

ARNHEIM 1972= M.T.W. Arnheim, The Senatorial Aristocracy in the Later Roman Empire, Oxford.

288

ARSLAN 1966 = E.A. Arslan, L’edificio termale romano detto “tempio di Castore e Polluce” presso Curinga (Catanzaro), Klearchos 8, 23-47.

ARSLAN 1974 = E.A. Arslan, Ville e città romane in Calabria, Magna Graecia IV, 9-10, 1-8.

ARSLAN 1990 = E.A. Arslan, La dinamica degli insediamenti in Calabria dal tardoantico al Medioevo, in CCARB 1990, 59-92.

ARSLAN 1991 = E.A. Arslan, Ancora da Scolacium a Squillace, in La Calabre 1991,465- 484.

ARSLAN 1998 = E.A. Arsian, Il territorio del Bruzio nel IV-V secolo (il paesaggio ruraIc), in Atti Taranto 1998, 391-430.

ARTHUR 1989 = P. Arthur, Some Observations on the Economy of Bruttium under the Later Roman Empire, JRA 2, 133-142.

ARTHUR 2004 = P. Arthur, From Vicus to Village: Italian Landscapes, AD 400-1000, in Christie 2004, 103-134.

ASHBY-LUGLI 1932 = Tb. Asby-G. Lugli, La basilica di Giunio Basso sull’Esquilino. RACr 9, 221-255.

Atti Taranto 1998 = L’ltalia meridionale in età tardoantica, Atti del 38° Convegno di Studi sulla Magna Grecia (Taranto 2-6 ottobre 1998), Taranto 1999.

AVETTA-MARCELLI-SASSO D’ELIA 1991 = L. Avetta-M. Marcelli-L. Sasso D’Elia, Quote S. Francesco, in La Calabre 1991, 559-609.

BACCI 2001 = G. Bacci (ed.), Patti Marina, il sito archeologico e l’Antiquarium, Messina.

BALDINI LIPPOLIS 1996 = I. Baldini Lippolis, Pavimenti in opus sectile nelle dimore di prestigio dell’Italia meridionale: alcuni esempi di età tardoantica, in Guidobaldi Guiglia Guidobaldi 1996, 653-664.

BALDINI LIPPOLIS 2001 = I. Baldini Lippolis, La domus tardoantica: forme e rappresentazioni dello spazio domestico nelle cittâ del Mediterraneo, (Studi e scavi 37), Boa logna.

BALSDON 1969= J.P.V.D. Baldson, Life and Leisure in Ancient Rome, New York.

BALMELLE 2001 = C. Balmelle, Les derneures aristocratiques d’, (Aquitania Suppl. 10), Paris.

289

BANAJI 2001 = J. Banaji, Agrarian Change in Late Antiquity. Gold, Labour, and Aristocratic Domination, Oxford.

BARELLO-CARDOSA 1991 = F. Barello-M. Cardosa, Casignana Palazzi, in La Calabre 1991, 669-687.

BARILLARO 1979 = E. Barillaro, La villa di Casinius a Casignana, Chiaravalle Centrale.

BARKER-LOYD-REYNOLDS 1985 = G. Barker, J. LOYD, J. Reynolds, J., (ed.), Cyrenaica in Antiquity , Oxford, 15-25.

BARNES 1982 = T.D. Barnes, The New Empire of Diocletian and Constantine, Cambridge MA.

BARNISH 1988 = S. Barnish, Transformation and Survival in the Western Senatorial Aristocracy, PBSR 66, 120-155.

BARTMAN 1988 = E. Bartman, Décor et Duplicatio: Pendants in Roman Sculpture Display, AJA 92 (1988).

BAUM-VOM FELDE 2003= P.C. Baum Vom Felde, Die geometrischen Mosaiken der Villa bei Piazza Armerina, Hamburg.

BECATTI 1951 = G. Becatti, Arte e gusto negli scrittori latini, Firenze.

BECATTI 1961 = G. Becatti (ed.), Mosaici e pavimenti marmorei, Scavi di Ostia, Ministero per i Beni Culturali e Ambientali, Vol. 4

BECKWITH 1970= J. Becwith, Early Christian and Byzantine Art, Yale.

BEDARD 1951 = W.M. Bedard, The Symbolism of the Baptismal Font in Early Christian Thought, Washington.

BEK 1983 = L. Bek, Questiones convivales. The Idea of the Triclinium and the Staging of Convivial Ceremony from Rome to Byzantium, ARID 12, 81-107.

BEKKER NIELSEN 1989 = T. Bekker Nielsen, The Geography of Power: Studies in the Urbanization of north-west Europe, Oxford.

BESCHACUACH, HANOUNE, THÉBERT 1977 = A. Beschacuach, R. Hanoune, Y. Thébert, Les ruines de Bulla Regia, Rome.

BIANCHI-BRUNO-DE NUCCIO 2002 = F. Bianchi-M. Bruno-M. De Nuccio, La Domus sopra le Sette Sale: la decorazione pavimentale e parietale dell’aula absidata, in De Nuccio-Ungaro 2002, 161-168.

290

BLAKE 1930 = M. E. Blake, The Pavements of the Roman Buildings of the Republic and Early Empire, MAAR, 8, 1930, 7-160.

BLANCHARD-LEMÉE 1996 et Alii = Mosaics of Roman Africa: Floor Mosaics from Tunisia, 1996.

BOËTHIUS-WARD PERKINS 1970 = A. Boëthius-J.B. Ward-Perkins, Etruscan and Roman Architecture, Harmondsworth.

BOLLA 1996 = M. Bolla, Le necropoli delle ville romane di Desenzano e Sirmione in Brogiolo 1996, 51-70.

BOUET 1997-98 = A. Bouet, Les thermes de la villa de Montmaurin (Haute-Garonne) et la pratique balnéaire et sportive dans l’Antiquite tardive, Aquitania 15, 2 13-244.

BOUET 2004 = A. Bouet, L’autocélébration des élites dans les thermes d’Occident au Haut- Empire, in Autocélébration des élites locales dans le monde romain. Contexte, texts, images (II s. av. J-C.-III s. ap. J.-C), M. Cebeillac-Gervasoni, L.Lamoine, and F.Trement (ed.)

BOWERSOCK-BROWN-GRABAR 1999 = G.W. Bowersock-P. Brown-O. Grabar (ed.), Late Antiquity. A Guide to the Postclassical World, Cambridge-London.

BOWES 2001 = K. Bowes, . . Nec sedere in villam: Villa Churches, Rural Piety and the Priscillianist Controversy, in T.S. Burns-J.W. Eadie (ed.), Urban Centers and Rural Contexts in Late Antiquity, East Lansing, 323-348.

BRENK 1999 = B. Brenk, La cristianizazione della domus dei Valerii sul Celio, JRA 33, 69-84.

BRENK 2003 = B. Brenk, Die Christianisierung der spatroniischen Welt. Stadt, Land, Haus, Kircher und Kioster in friihchristlicher Zeit, (Studien und Perspectiven 10), Wiesbaden.

BROGIOLO 1996 = G. P. Brogiolo, La fine delle ville romane: trasformazioni nelle campagne tra tarda antichità e alto medioevo, Mantova.

BROGIOLO-GAUTHIER-CHRISTIE 2000 = G.P. Brogiolo-N. Gauthier-N. Christie (ed.), Towns and their Territories between Late Antiquity and the Early Middle Ages, (The Transformation of the Roman World 9), Leiden-Boston-Köln.

BROGIOLO-WARD-PERKINS 1999 = G.P. Brogiolo-B. Ward-Perkins (ed.), The Idea and Ideal of the Town between Late Antiquity and the Early Middle Ages, (The Trasformation of the Roman World 4), Leiden.

BROWN 1981 = P. Brown, The cult of the saints, Chicago.

291

BROWN 1987 = P Brown, Late Antiquity, in P. Veyne, A. Goldhammer (ed.), A History of Private Life: I. From Pagan Rome to Byzantium.

BROWN 1988 = P. Brown, The Body and Society, New York.

BROWN 1992 = P. Brown, Power and Persuasion in Late Antiquity: Towards a Christian Empire, Madison, Wis.

BROWN 2002 = P. Brown, Poverty and leadership in the later Roman Empire, Hannover

BURRUS 2005 = V. Burrus (ed.), Late Ancient Christianity, Minneaopolis.

CAGIANO DE AZEVEDO 1986 = M. Cagiano de Azevedo, Casa, città e campagna nd Tar-do Antico e nell’Alto Medioevo, C.D. Fonseca-D. Adamesteanu-F. D’Andria (ed.), (Archeologia e Storia 1), Galatina.

CAMERON 1986 = A. Cameron, Redrawing the map: early Christian territory after Foucault, in Journal of Roman Studies 76 (1986), 266-271.

CAMERON 1993 = A. Cameron, The Mediterranean World in Late Antiquity, London- New York.

CAMERON 1999 = A. Cameron, Il tardo impero romano, Bologna (english edition: Cambridge MA 1993).

CANTARELLI 1984= F. Cantarelli, Dionisio di Alicarnasso. Storia di Roma arcaica (Le antichità romane), Milano.

CANTINO WATAGHIN 1994 = G. Cantino Wataghin, Tardoantico e altomedioevo nel territorio padano. Il territorio, in Francovich-Noyé 1994, 142-147.

CANTINO WATAGHIN 2000 = G. Cantino Wataghin, Christianisation et organisation ecclesiastique des campagnes: l’Italie dii Nord aux IV-VI siècles, in Brogiolo-Gauthier Christie 2000, 209-234.

CANTINO WATAGHIN-UGGÈ 2001 = G. Cantino Wataghin-S. Uggè, Scavi e scoperte di archeologia cristiana in Italia settentrionale (1993-1998), in Gandolfi 2001,1,7-38.

CAPOGROSSI COLOGNESI 1982 = L. Capogrossi Colognesi, L’agricoltura romana. Guida storica e critica, Roma.

CAPOGROSSI COLOGNESI 2002 = L. Capogrossi Colognesi, Persistenza e innovazione nelle strutture territoriali dell’Italia romana. L’ambiguità di una interpretazione storiografica e dei suoi modelli, Napoli.

292

CARANDINI-RICCI-DE VOS 1982 = A. Carandini-A. Ricci-M. de Vos, Filosofiana. La villa di Piazza Armerina, Palermo.

CARANDINI 1983 = A. Carandini, Dopo una prima fase del dibattito su Filosofiana, in AA.VV. 1983, 595-602.

CARANDINI 1993 = A. Carandini, Paesaggi agrari meridionali ed etruschi a confronto, Atti del Convegno per il Bimillenario della morte di Q. Orazio Flacco (Venosa 1992), Venosa, 239-245.

CARANDINI 1995 = A. Carandini, Il latifondo in epoca romana, fra Italia e province, in Du Latifundium au latifondo, Un héritage de Rome, une creation médiévale ou modèrne, Actes de la table ronde (Bordeaux 1992), Paris, 31-36.

Carranque 2001 = Carranque, Centro de Hispania romana, (Museo Arqueologico Regional Alcalá de Henares, 27 de abril a 23 de septiembre 2001), Madrid.

CARCOPINO 1940 = J. Carcopino, Daily life in ancient Rome; the people and the city at the height of the empire, New Haven and London.

CARRIÉ-LIZZI TESTA 2002 = J.-M. Carrie, R. Lizzi Testa (ed.), “Humana sapit”, Etudes d’AntiquitC Tardive offertes a Lellia Cracco Ruggini (Bibliothèque de 1’Antiquite Tardive 3), Paris.

CECCONI 1994= G. A. CECCONI, Governo imperiale e élites dirigenti nell'Italia tardoantica : problemi di storia politico-amministrativa,Como.

CECCONI 2006= G.A.CECCONI, “Honorati, possessores, curiales: competenze istituzionali e gerarchiche di rango nella città tardoantica,” in Le trasformazioni delle élites in età tardoantica, RITA LIZZI TESTA (ed.), Roma 2006, pp.41-64.

CCARB 40, 1993= XL corso di cultura sull'arte ravennate e bizantina: Primo Colloquio dell'Associazione Italiana per lo Studio e la Conservazione del Mosaico (AISCOM): seminario internazionale di studi su : "La Albania dal Tardoantico al Medioevo, aspetti e problemi di Archeologia e Storica dell'arte" : Ravenna, 29 aprile-5 maggio 1993, Ravenna.

CHASTAGNOL 1963 = A. Chastagnol, L’administration du Diocése italien au Bas- Empire,Historia 12, 348-379.

CHASTAGNOL 1977 = A. Chastagnol, Le problème du domicile légal des sénateurs romains à l’epoque impériale, in Mélanges offerts à Leopold Sédar Senghor, Dakar, 43-54.

293

CHASTAGNOL 1982 = A. Chastagnol, La carriére senatoriale du Bas Empire (depuis Dioclétien), in Epigrafia e ordine senatorio, Atti del Colloquio internazionale, AIEGL 1, Roma,167-194.

CHASTAGNOL 1992 = A. Chastagnol, Le Sénat romain à l’ époque imperial: recherches surla composition de l’assemblée et le slatut de ses membres, Paris.

CHAVARRIA ARNAU 2004 = A. Chavarria Arnau, Interpreting the Transformation of Late Roman Villas: The Case of Hispania, in Christie 2004, 67-102.

CHAVARRIA-LEWIT 2004 = A. ChavarrIa-T. Lewit, Archaeological Research on the Late Antique Countryside: a Bibliographical Essay, in Lavan-Bowden-Machado 2004,3-52.

CHAVARRIA-ARCE-BROGIOLO 2006= A. Chavarria, J. Arce, G.P. Brogiolo, Villas Tardoantiguas en el Mediterráneo Occidental, Madrid.

CHRISTIE 2004 = N. Christie (ed.), Landscapes of Change in Late Antiquity and the Early Middle Ages: Themes, Directions and Problems, Ashgate.

CLARK 1984 = E.A. Clark, The Life of Melania the Younger (Studies in Women and Religion 14), New York and Toronto.

CLARKE 1979 = J.R. Clarke, Roman Black and White Figural Mosaics.

CLARKE 1991= J. R. Clarke, The Houses of Roman Italy, 100 BC- AD 250: Ritual, Space, and Decoration, Berkeley.

CME = Corpus de Mosaicos de España, Madrid 1978-.

CMGR I = La mosaïque Gréco-romaine. Colloques internationaux du Centre National de la Recherche Scientifique, Paris, 1963 (Paris 1965).

CMGR VI = VI Colloquio International sobre Mosaico antiguo, Palencia-Mérida, Octubre 1990 (Guadalajara 1994).

COLICELLI 1995 = A. Colicelli, Gli insediamenti di eta romana nei Bruttii: un nuovo censimento (1991-1995), ArchStCalLuc 62, 47-95.

CONNOLLY DODGE 1998= P. Conolly, H. Dodge, La Ciudad Antigua. La vida en la Atenas y Roma clásicas, Madrid.

CORBO 1969= V.C. Corbo, The house of St. Peter at Capharnaum; a preliminary report of the first two campaigns of excavations, April 16-June 19, Sept. 12-Nov. 26, 1968, Jerusalem.

CORSI 2000 = C. Corsi, Le strutture di servizio del cursus publicus in Italia: ricerche topografiche ed evidenze archeologiche, Oxford.

294

COSTABILE 1976 = F. Costabile, Municipium Locrensium, Reggio.

COSTAMAGNA-SABBIONE 1990 = L. Costamagna-C. Sabbione, Una città in Magna Grecia, Locri Epizefiri, Reggio Calabria.

COURCELLE 1938 = P. Courcelle, Le site du monastère de Cassiodore, MEFR 55, 259- 307.

CRACCO RUGGINI 1961 = L. Ruggini, Economia e società nell’Italia annonaria: rapporti fra agricoltura e commercio dal IV al Vi secolo d.C., (Fondazione Guglielmo Castelli 30), Milano.

CRACCO RUGGINI 1998 = L. Cracco Ruggini, Il senato fra due crisi (III- VI secolo), in Il Senato nella storia, Il senato netl’età romana, Roma, 223-375.

CRACCO RUGGINI 2004 = L. Cracco Ruggini, The Italian City from the Third to the Sixth Century: “Broken History” or everchanged Kaleidoscope?, in Straw, Lim 2004, 33-48.

CRAWFORD 1996 = M.Crawford, Roman Statutes, II, London.

CROWTHER 2007 = N.B. Crowther, Sport in Ancient times, Westport, Conn.

D’ARMS 1970 = J.H. D’Arms, Romans on the Bay of Naples. A Social and Cultural Study of the Villas and their Owners from 150 BC to AD 400, Cambridge.

DE ANGELIS BERTOLOTTI-IOPPOLO-PISANI-SARTORIO 1988 = R. De Angelis Bertolotti-G. Toppolo-Pisani-G. Sartorio, La residenza imperiale di Massenzio. Villa, mausoleo e circo, Rome.

DE BLOIS 2002 = L. De Blois, The Crisis of the Third Century A.D. in the Roman Emei pire: A Modern Myth???, in L. De Blois, J. Rich, The Transformation of Economic Lf under the Roman Empire, Proceedings of the Second Workshop of the International Network, Impact of Empire (Roman Empire, c. 200 B.C.-A.D. 476) (Nottingham, July 4-7 2001), Amsterdam, 204-217.

Décor I = que romaine. , Paris 1985. Décor II = que romaine. , Paris 2002.

DE FRANCESCHINI 1991= M. De Franceschini,Villa Adriana. Mosaici-pavimenti-edifici, Rome.

DE FRANCISCIS 1988 = A. De Franciscis (ed), La villa romana del Naniglio di Gioiosa

295 lonica, Relazione preliminare delle campagne di scavo 1981—1986 dirette da Alfonso De Franciscis, (Archaia, Monumenti 1), Napoli.

DE NITTIS 2006 = V. De Nittis, La villa romana di Casignana. I balnea, l’aula basilicale e la galleria frontale tra due torri, in , Studi interdisciplinari sul mondo antico, 2, Roma.

DE NITTIS 2007= V. De Nittis, Forme e simboli nell’architettura della villa, in La villa romana di Palazzi di Casignana. Guida archeologica, Gioiosa Jonica.

DE NUCCIO-UNGARO 2002 = M. De Nuccio, L. Ungaro, I marmi colorati della Roma imperiale, (Roma, Mercati di Traiano 28 settembre 2002 - 19 gennaio 2003), Venezia.

DI PAOLA 1999 = L. Di Paola, Viaggi, trasporti e istituzioni: studi sul cursus publicus, Messina.

DIAZ 2000 = P.C. Diaz, City and Territory in Hispania in Late Antiquity, in BrogioloGauthier- Christie 2000, 3-36.

DOLEGER 1934 = F.J. Doleger, “Zur Simbolik des altchristlichen Taufhauses,” AC 4 (1934), 153-187.

DUNBABIN 1978 = K.M.D. Dunbabin, The Mosaics of Roman North Africa: Studies in Iconography and Patronage, (Oxford Monographs on Classical Archaeology 16), Oxford.

DUNBABIN 1983 = K.M.D. Dunbabin, The San Giovanni Mosaics in the Context of Late Roman Mosaic Production in Italy, in Gualtieri-Salvatore-Small 1983, 47-57.

DUNBABIN 1991 = K.M.D. Dunbabin, Triclinium and Stibadium, in W.J. Slater (ed), Dining in a Classical Context, Ann Arbor, 12 1-148.

DUNBABIN 1999 = K,M.D. Dunbabin, Mosaics of Greek and Roman World, Cambridge.

DUNBABIN 2003 = K.M.D. Dunbabin, The Roman Banquet, Images of Conviviality, Cambridge.

DUPONT 1992 = F. Dupont, Daily Life in Ancient Rome, Oxford.

DUVAL 1960 = N. Duval, Que savons nous du palais de Théodoric a Ravenne?, MEFRA 77, 369-371.

DUVAL 1984 = N. Duval, Les maisons d’Apamée et l’architecture “palatiale”de l’Antiquite Tardive, in J. Balty (ed.), Apamée de Syrie, Bilan des recherches archeologiques 1973-79, Aspects de 1 ‘architecture domestique d’Apamée, Actes du Colloque (Bruxelles 29-31 mai 1980), Bruxelles, 447-470.

296

DUVAL 1986 = N. Duval, L’iconographie des “villas africaines” et la vie rurale dans l’Afrique romaine de l’Antiquité tardive, in Histoire et Archéologie de l’Afrique du Nord, III Colloque sur l’Histoire et l’Archéologie de l’Afrique du Nord, III Congr. Nat. Soc. Say., (Montpellier 1985), Paris, 163-176.

DUVAL 1987 = N. Duval, Existe-t-il une “structure palatiale” propre a l’Antiquite Tardive?, in Le système palatiale en Orient, en Grèce et a Rome, Actes du Colloque (Strasbourg 19-22 Juin 1985), Strasbourg, 463-489.

DUVAL 2002 = N. Duval, Deux mythes de l’iconographie de l’antiquité tardive: la villa fortifiée et le “chasseur vandale,” in Carrié-Lizzi Testa 2002, 333-340.

DYGGVE 1945 = E. Dyggve, Recherches sur le palais imperial de Thessalonique.

ELLIS 1985 = S. Ellis, The “Palace of the Dux” at Apollonia and Related Houses, in G. Barker-J. Lloyd-I. Reynolds (eds.), Cyrenaica in Antiquity, Oxford (BAR mt. Ser. 236), 15-25.

ELLIS 1988= S. Ellis, The End of the Roman House, AJA 92, 565-576.

ELLIS 1991 = S. Ellis, Power, Architecture, and Decor: How the Late Roman Aristocrat Appeared to His Guest, in E.K. Gazda (ed.), Roman Art in the Private Sphere: new Perspective on the Architecture and Decor of the domus, villa et insula, Ann Arbor 1991, 117-135.

ELLIS 1995 = S. Ellis, Classical Reception Rooms in Romano-British Houses, Britannia 26, 1995, 163-178.

ELLIS 1997 = S. Ellis, Late-antique Dining: Architecture, Furnishings and Behaviour, in Laurence-Waflace-Hadrjll 1997, 41-52.

ELLIS 2000 = S. Ellis, Roman Housing, London.

ELLIS 2007= S. Ellis, Sheding Light on Late Roman Houses, in Housing in Late Antiquity: from Palaces to Shops, L. Lavan et alii (ed.), Boston.

ELLIS- KIDNER 2004 (ed.) = L. Ellis and F. Kinder, Travel, Communication and Geography in Late Antiquity.

ETIENNE 1992 = R. Etienne, Ausone, propriétaire terrier ei Ic problème du latifttndium au IV siècle ap. J.C. in M. Christol-S. Demougin-Y. Duval-C. Lepelley-L. Pietri (ed.), Institutions, Société et vie politique dans l’empire romain au IV siècle ap.J.C., Actes de la table ronde autour l’ouvre d’André Chastagnol, (Paris 20-21 janvier 1989), Paris, 305-311.

ETIENNE-DE ALERCÂO-MAYET 1989 = R. Etienne-J. de Alercâo-F. Mayet, Les villas romaines de Sâo Cucufate (Portugal), CRAI, 232-249.

297

FAGAN 1992= G. Fagan, Bathing in Public in the Roman World, Ann Arbor.

FAGAN 2005 = G. Fagan, Thebert, Y., Thermes romains d'Afrique du nord et leur contexte Mediterraneen: etudes d'histoire et d'archeologie, Journal of Roman Studies 95, 2005, 318-319

FARIOLI 1982 = F. Farioli, La trichora di S. Martino e il monastero Vivariense seu Castellense di Cassiodoro, in M.L. Gualandi-L. Massei-S. Settis (ed.), APARCHAI. Nuove ricerche e studi sulla Magna Greica e la Sicilia antica in onore di Paolo Enrico Arias, Pisa, 669-677.

FENDRI 1963 = M. Fendri, Découvertes archéologiques dansl la région de Sfax. La mosaïque des Océans, Tunis.

FERNÁNDEZ CASTRO 1982 = M.C. Fernández Castro, Villas romanas en España, Madrid.

FERNÁNDEZ-GALIANO1999= D. Fernández-Galiano, “La villa romana de Carranque,” in Hispania. El legado de Roma. Madrid.

FIACCADORI 1994 = F. Fiaccadori, Calabria tardoantica, in Settis 1994, 700-762.

FLEISCHER-LUN-NIELSEN 2001 = J. Fleischer, J., Lund, M. Nielsen (ed.), Late Antiquity: art in context, Copenaghen.

FOUCAULT 1986 = M. Foucault, The Care of the Self, New York.

FOUCHER 1961 = L. Foucher, Découvertes archéologiques à Thysdrus eb 1961, Tunis.

FOUCHER 1963 = L. Foucher, La Maison des masques à Sousse, Fouilles 1962-1963, Tunis.

FRANCOVICH—NOYÉ 1994 = R. Francovich and G. Noyé (eds.), Atti del Convegno. La storia dell’Alto Medioevo italiano (VI-IX secolo) alla luce dell’archeologia (Siena 1992), Firenze.,

GAMBI 1965= L. Gambi, La Calabria, Torino.

GARBARINO 1988 = P. Garbarino, Ricerche sulla procedura di ammissione al senato nel tardo impero romano, (Memorie dell’Istituto Giuridico, Univ. di Torino, serie III, 26), Milano.

GAUCKLER 1913 = P. Gauckler, Basiliques chrétiennes de Tunisie, Paris.

GENTILI 1999= G.V. Gentili, La villa romana di Piazza Armerina, Palazzo Erculio, 1-111, Osimo.

GERMAIN 1969 = S. Germain, Les mosai , Paris.

298

GERVASONI-LAMOINE-TRÉMENT 2004 = M. c-Gervasoni, L. Lamoine, F. Trément, : contextes, images, textes (IIe s. av. J.-C.-IIIe s. ap. J.-C.).

GHISLANZONI 1962 = F. Ghislanzoni, La villa romana in Desenzano, Milano.

GIARDINA 1982 = A. Giardina, Lavoro e storia sociale: antagonismi e alieanze dall’ellenismo altardoantico, Opus 1, 115-146.

GIARDINA 1986a A. Giardina (ed), Società romana e impero tardoantico, I-IV, Roma Bari.

GIARDINA 1986b= A. Giardina, Le due Italie nella forma tarda dell’impero, in Giardina 1986 a, I, 1-30.

GIARDINA 1986c = A. Giardina, Palladio, il latifondo italico e l’occultazione della società rurale, Appendice alle due Italie, in Giardina 1986a, I, 3 1-36.

GIARDINA 1986d = A. Giardina, Carità eversiva. Le donazioni di Melania la Giovane e gli equilibri della società tardoaromana, Studi Tardoantichi 2, 77-102.

GIARDINA 1993 = A. Giardina, La formazione dell’Italia provinciale, in Carandini-Cracco Ruggini-Giardina 1993, 1, 5 1-68.

GIARDINA 1994 = A. Giardina, Melania, la santa, in A. Fraschetti (ed.), Roma al femminile, Roma-Bari, 259-285.

GIARDINA 1997 = A. Giardina, L’Italia, il modo di produzione schiavistico e i tempi di una crisi, in A. Giardina, L’Itaiia romana. Storia di an ‘identità incompiuta, Roma- Bari, 233-264.

GIARDINA 1998 = A. Giardina, Considerazioni finali, in Atti Taranto 1998, 609-624.

GIARDINA 1999 = A. Giardina, Esplosione di tardoantico, in Mazzoli-Gasti 1999, 9-30.

GIARDINA-SCHIAVONE 1981 A. Giardina-A. Schiavone (ed.), Società romana e produzione schiavistica, I-III, Roma-Bari.

GINOUVÈS 1955 = R. Ginouvès, Sur un aspect de l'évolution des bains en Grèce vers le 4ème Siècle de notre ère in Bulletin de Correspondance Hellénique 79.

GIVIGLIANO 1994 = G.P. Givigliano, Percorsi e strade, in Settis 1994, 243-337.

GNOLI 1971 = R. Gnoli, Marmora Romana, Rome.

GOODCHILD 1976 = R. G. Goodchild, The Palace of the Dux, in Goodchild et Alii 1976, 345- 276.

299

GORDON 1983= A. E. Gordon, Illustrated Introduction to Latin Epigraphy.

GORGES 1979 = J.-G. Gorges, Les villas hispano-romaines. Inventaire et problématique archéologiques, (Publications du Centre Pierre Paris 4), Paris.

GORGES 1990 = J.-G. Gorges, Villes et villas de Lusitanie (Interactions-échanges-autonoinies), in Les villes de Lusitanie romaine. Hierarchies et territoires, Table ronde internationale dii CNRS (Talence, 8-9 december 1988), Paris, 91-113.

GRAHAM 2006= M.W. Graham, News and frontier consciousness in the late Roman Empire, Ann Arbor.

GREEN 1991 = R. Green, Works of Ausonius, Oxford.

GRELLE1998= F. Grelle, “Ordinamento provinciale e organizzazione locale nell’Italia Meridionale” in L’Italia Meridionale in Età Tardoantica. Atti del 38esimo Convegno di Studi sulla Magna Grecia, Taranto 2-6 Ottobre, pp.115-139.

GRELLE-VOLPE 1996 = F. GreIle-G. Volpe, Aspetti della geografia amministrativa ed economica della Calabria in eta tardoantica, in M. Pani (ed.), Epigrafia e territorio. Politica e società. Teini di antichità romane, IV, Barin, 113-155.

GRELLE 1998 = F. Grelle, Ordinamento provinciale e organizzazione locale nell’italia meridionale, in Atti Taranto 1998, 115-139.

GREY 2007 = C. Grey, “Contextualizing Colonatus: The Origo of the Late Roman Empire,” in Journal of Roman Studies 97 (2007), pp.155-175

GUALTIERI 1994 = M. Gualtieri, La villa romana di Masseria Ciccotti (Oppido Lucano,PZ): primi dati sul paesaggio rurale dell’Alto Bradano in età imperiale, Bollettinostorico della Basilicata 10,49-73.

GUALTIERI 2003 = M. Gualtieri, La Lucania romana. Cultura e società nella documentazione archeologica, (Quademi di Ostraka 8), Napoli.

Guida = La villa romana di Palazzi di Casignana. Guida archeologica, Sabbione C. (ed.), Gioiosa Jonica, 2007.

GUIDOBALDI 1986 = F. Guidobaldi, L’edilizia abitativa unifamiliare nella Roma tardoantica, in Giardina 1986 a, II, 165-234.

GUIDOBALDI 1993 = F. Guidobaldi, Roma. II tessuto abitativo, le “domus”, i “tituli”, in Carandini-Cracco Ruggini-Giardina 1993, 2, 69-83.

GUIDOBALDI 1994- F. Guidobaldi, Sectilia Pavimenta di Villa Adriana, (MosAntIt, Studi Monografici, Rome

300

GUIDOBALDI 1999 = F. Guidobaldi, Le domus tardoantiche di Roma come “sensori” delle trasformazioni culturali e sociali, in W.V. Harris (ed.), The Transformations of Urbs Roma in Late Antiquity, (JRA Suppi. Series 33), Portsmouth, 53-68.

GUIDOBALDI 2000 = F. Guidobaldi, Distribuzione topografica, architettura e arredo delle domus tardoantiche, in Ensoli-La Rocca 2000, 134-136.

GUIDOBALDI 2004= .Guidobaldi, Le residenze imperiali della Roma tardoantica, in Balmelle- Chevalier-Ripoll 2004, 37-46.

HALES 2004 = S. Hales, The Roman House and Social Identity, Cambridge.

HANSEN 1997 = S. L. Hansen, The embellishment of Late-antique domus in Ostia and Rome, in Patron and Pavements in Late Antiquity, S. Isager-B. Poulsen (ed.), Odense.

HAYES 1972 = J. Hayes, Late Roman Pottery, London.

HEATHER 1998 = P. Heather, Senators and Senate, III, A. Cameron-P. Garnsey (ed.), The Cambridge Ancient History’, 13, The Late Empire AD. 337-425, Cambridge.

HEATHER 2006= The Fall of the Roman Empire: A New History of Rome and the Barbarians, Oxford.

HIDALGO PRIETO 1996 = R. Hidalgo Prieto, Espacio publico y espacio privado en el conjunto palatino di Cercadilla (Córdoba): el aula central y las termas, Sevilla.

HINGLEY 1989 = R. Hingley, Rural Settlement in Roman Britain.

HIRSCHFELD 1999 = Y. Hirschfeld, Habitat, in Bowersock-Brown-Grabar 1999, 258-272.

HOBSON 2009 = B. Hobson, Latrinae et foricae: toilets in the Roman world, London.

HORDEN-PURCELL 2000= P. Horden, N. Purcell, The Corrupting Sea: A Study of Mediterranean History, Malden, Mass.

HUMBERT 2005 = M. Humber (ed.), Le Dodici Tavole. Dai Decemviri agli Umanisti, Pavia.

HUMPHREY- OLESON-SHERWOOD 1998 = J.W. Humphrey, J.P. Oleson, A.N. Sherwood, Greek and Roman technology: A SourcebooK.

IACOPI 1993= I. Iacopi, Domus Aurea di Nerone, in “Functional and Spatial Analysis of Wall Painting,” Proceeding of the Fifth International Congress on Ancient Wall Painting (Amsterdam 8-12 September 1992), Leiden.

301

IOPPOLO-PISANI SARTORIO 1999 = G. Ioppolo-G. Pisani Sartorio (ed.), La villa di Massenzio sulla via Appia, Il circo, Roma.

IRELAND 1984 = R. I. Ireleand, Anonymi Auctoris De Rebus Bellicis, Bibliotheca Scriptorvm Graecorvm et Romanorvm Tevbneriana, Lipsiae.

JACOBY 1905= C. Jacoby, Dionysii Halicarnasei Antiquitatum romanarum quae supersunt, IV, Lipsiae.

JONES 1964 = A.H.M. Jones, The Later Roman Empire, 284-602: A Social Economic and Administrative Survey, Oxford.

JONES 1974 = A.H.M Jones, The Roman Economy. Studies in Ancient Economic and Administrative History, Oxford.

KEAY 1988= S. J. Keay, Roman Spain, London.

KEHOE 2007 = D. P. Kehoe, Law and Rural Economy in the Roman Empire, Ann Arbor.

KELLY 2004 = C. Kelly, Ruling the Later Roman Empire, Cambridge, MA.

KHADER 2001 = Aicha Ben Abed Ben Khader et alii, Recherches franco-tunisiennes sur la mosai , Ecole francaise de Rome, Institut national du Patrimoine (Tunis), Rome 2001.

KAHRSTEDT 1960= U. Kahrstedt, Das wirtschaftliche Gesicht Griechenlands in der Kaiserzei: Kleinstadt, Villa und Domane, Wiesbaden.

KRAUTHEIMER 1942 = R.Krautheimer, Introduction to an ‘Iconography’ of Mediaeval Architecture, in Journal of the Warburg and Courtauld Institutes, V, 1942, pp. 1-33,

KRAUTHEIMER 1986 = R. Krautheimer, Early Christian and Byzantine Architecture, Harmondsworth (4th ed).

KRENCHER-KRÖGER 1929 = D. M. Krencker-E. Kroger, Die Trierer Kaisertherinen, Augsburg.

KUZISCIN 1981= V. I Kuziscin, L’organizzazione del latifondo in Italia alla fine della repubblica, in CAPOGROSSI COLOGNESI, L. (ed.), L’agricoltura romana, Roma.

LA GUARDIA 1994 = R. La Guardia (ed.), Studi sulla villa romana di Desenzano, 1, Milano.

LANÇON 1999 = B. Lançon, La vita quotidiana a Roma nel tardo impero, Milano.

LANÇON 2000 = B. Lançon, Rome in Late Antiquity. Everyday Life and Urban Change, AD 312-609, New York.

302

LATTANZI 1973 = E. Lattanzi, La villa rornana di Porto Saturo presso Taranto, Cenacolo III, 1-3, 43-48.

LAURENZI 2004 = E. Laurenzi, Edifici di spettacolo in contesti privati: Ia villa di Pianosa, in Burri-Delacrétaz-Monnier-Nobili 2004, 188-199.

LAVAN 2001 = L. Lavan (ed.). Recent Research in Late-antique Urbanism, JRA Suppl. Series 42.

LAVAN-BOWDEN-MACHADO 2004 = L. Lavan- W.Bowden- C. Machado (ed.), Recent Research on the Late Antique Countryside, (Late Antique Archaeology 2), Leiden.

LE MAHO 1994 = J. Le Maho, La réutilisation funéraire des edifices antiques en Normandie au cours du Haut Moyen Age, in Fixot-Zadora Rio 1994, 10-21.

LEACH 1997 = E. W. Leach, Oecus on : investigating the vocabulary of the Roman house, in S. Bon – R. Jones (ed.), Sequence and Space in Pompei, Oxford, 50-72.

LEANZA 1986 = S. Leanza (ed), Flavio Magno Aurelio Cassiodoro, Atti della Settimana di Studi (Cosenza-Squillace 19-24 settembre 1983), Catanzaro.

LEE 1999 = D. Lee, Pagans and Christians. A Sourcebook, London and New York.

LEPORE 1989 = E. Lepore, Origini e strutture della Campania antica, Bologna.

LEVI 1947 = D. Levi, Antioch Mosaic Pavements, 2 vols. Princeton.

LÉZINE 1969 = A. Lezine, Carthage-Utique. Etudes d’architecture et d’urbanisme, Paris 1969.

LEWIT 1991 = T. Lewit, Agricoltural Production in the Roman Economy AD 200-400, (BAR Inter. Ser. 568), Oxford.

LIEBESCHUETZ 2001 = J.H.G.W. Liebeschuetz, The Use and Abuses of the Concept of “Decline” in Later Roman History, or was Gibbon Politically incorrect?, in Lavan 2001, 233-245.

LIEBESCHUETZ 2001a = J. H. G. W. Liebeschuetz, The Decline and Fall of the Roman City, New York.

LING 1998 = R. Ling, Ancient Mosaics, Princeton.

LITTLEWOOD 1987 = A .R. Littlewood, Ancient Literary Evidence far the Pleasure Gardens of Roman Country Villas, in E. Blair MacDougall (ed), Ancient Roman Villa Gardens, Dumbarton Oaks Colloquiwn on the History of Landscape Architecture 10th, (1984), Washington, 7-30.

303

LO CASCIO 1997 = E. Lo Cascio (ed.), Terre, proprietari e contadini dell’impero romano: dall’affitto agrario al colonato tardoantico, (Ricerche - Ia Nuova Italia Scientifica 16), Roma.

LO CASCIO-STORCHI MARINO 2001= E. Lo Cascio, D. Storchi Marino (ed.), Modalità insediative e strutture agrarie nell’Italia meridionale in eta romana, Atti del Convegno internazionale (Napoli ]1-13 giugno 1998), (Pragmateiai 7), Bari.

LONGOBARDI 2006 = N. Longobardi (ed.), In stabiano: exploring the ancient seaside villas of the roman elite.

LUSCHI 1982 = L. Luschi, Basilicae centenariae tres in Studi classici e orientali, 32, 157-178.

McKAY 1975 = AG. McKay, Houses, Villas and Palaces in the Roman World, London.

McMULLEN 1981 = R. MacMullen, Paganism in the Roman Empire, New Haven.

McMULLEN 1997 = Christianity and Paganism in the Fourth to Eighth Centuries, New Haven.

MACKIE 2003 = G. Mackie, Early Christian Chapels in the West, Decoration, Function, and Patronage, Toronto-Buffalo-London.

MALMBERG 2003 = S. Malmberg, Dazzling Dining. Banquets as an Expression of Imperial Legitimacy, Uppsala.

MARASOVIC-MARASOVIC 1994 = J. Marasovic-T. Marasovic, Le ricerche nel palazzo di Diocleziano a Split negli ultimi 30 anni (1964-1994), AnTard 2, 89-106.

MARCONE 1983 = A. Marcone, Commento storico al libro VI dell’Epistolario di Q. Aurelio Simmaco, (Biblioteca di Studi antichi 37), Pisa.

MARCONE 1988 = A. Marcone, Il colonato romano nella storiografia moderna (da Fustel de Coulanges ai nostri giorni), (Biblioteca di Athenaeum 7), Como.

MARCONE 1992-93 = A. Marcone, Costantino e l’aristocrazia pagana di Ronw, in G. Bonamente-F. Fusco (ed.), Costantino ii Grande, II, (Macerata 18-20 dicembre 1990), Macerata, 645-658.

MARCONE 1993 = A. Marcone, Ii lavoro neile campagne, in Carandini-Cracco RugginiGiardina 1993, 823-844.

MAZZARINO 1951 S., Aspetti Sociali del Quarto Secolo,

MARTIN 1971 = R. Martin, Recherches sur les agronomes latins et leurs conceptions économiques et sociales, Paris.

304

MARTIN 1997 = R. Martin, La villa romana de Torre Águila a partir del siglo IV d.C. Considerationes generales, Congreso Internacional. La Hispania de Teodosio, Segovia, 1997, pp. 607-711.

MASSARI-ROFFIA-BOLLA-CAPORUSSO 1985 = G. Massari-E. Roffia-M. Bolla-D. Caporusso, La villa tardoromana di Palazzo Pignano (Cremona), in G. Pontiroli (ed.) Crernona romana, Atti del Congresso storico-archeologico per il 22000 anno difondazione di Cremona (Cremona 30-31 nuiggio 1982), Cremona, 185-259.

MATTHEWS 1975 = J. Matthews, Western Aristocracies and Imperial Court, AD 364 -425, Oxford.

MAZZARINO 1951 = S. Mazzarino, Aspetti sociali del quarto secolo: ricerche di storia tardo- romana, (Problemi e ricerche di stona antica 1), Roma.

MAZZARINO 1972= S. Mazzarino, Si puó dire ‘Bruttium’? La denominazione tardo romana dell’attuale Calabria, in Archeologia Storica Pugliese, 25, 1972, pp.463-467.

MEATES 1987 = G.W. Meates, The Roman Villa at Lullingstone, Kent, Volume II: The Wail Paintings and Finds, Maidstone.

MERTEN 1983 = E. W. Merten, Bäder und Badegepflogenheiten in der Darstellung der Historia Augusta, Bonn.

MIELSCH 1990 = H. Mielsch, La villa romana, Firenze (ed. orig. München 1987).

MILLER 1962 = K. Miller, Die Peutingerische Tafel, Stuttgart.

MICKWITZ 1959 = G. Mickwitz, Marketing and Competition. The various Forms of Competition at the Successive Stages of Production and Distribution, Helsingfors.

MINUTO 1990 = D. Minuto, Chiese e monasteri in Calabria dal Tardoantico all’Alto Medioevo, in L’Italia meridionale tra Goti e Longobardi, CCARB 1990, 303-366.

MIRABELLA ROBERTI 1965 = M. Mirabella Roberti, Una basilica paleocristiana a Palazzo Pignano, Insula Fulcheria 4, 79-90.

MIRABELLA ROBERTI 1968 = M. Mirabella Roberti, Ancora sulla “Rotonda” di Palazzo Pignano, Insula Fuicheria 7, 85-94.

MIRABELLA ROBERTI 1969 = M. Mirabella Roberti, Scoperto ii Palatium di Palazzo Pignano, Insula Fuicheria 8, 19-23.

MIRABELLA ROBERTI 1971 = M. Mirabella Roberti, Le ricerche di archeologia cristiana in Lombardia dal 1958-1968, in Atti del II Congresso Nazionale di Archeologia Cristiana (Matera 25-31 maggiol969), Roma, 337-354.

305

MIRABELLA ROBERTI 1994 = M. Mirabella Roberti, Nuovi mosaici dalla villa romana di Desenzano, in J.-P. Darmon, A. Rebourg (ed.), La mosaïque greco-romaine, IV colloque international pour l’étude de la mosaïque antique (Trèves 8-14 aotit 1984), Paris, 107-114.

MORVILLEZ 1995 = E. Morvillez, Les sailer de reception triconques dans l’architecture domestique de l’Antiquité tardive en Occident, Histoire de l’art 31, pp. 15-26. MORVILLEZ 1996 = E. Morvillez, Sur les installations de lits de table en sigma clans l’architecture domestique du Haut et du Bas-Empire, Pallas 44, 119-158.

NEUDECKER 1988 = R. Neudecker, Die Skulpturenausstattung romischer Villen in Italien, Mainz am Rhein.

NEUDECKER 1994 = R NEUDECKER, Die Pracht der Latrine. Zum Wandel öffentlicher Bedürfnisanstalten in der kaiserzeitlichen Stadt. Studien zur antiken Stadt, .

NIELSEN 1990= I. Nielsen, Thermae et balnea. I. Text, Aarhus University Press.

NOVAK 1979 = D.M. Novak, Constantine and the Senate: an Early Phase of the Christianisation of the Roman Aristocracy, Ancient Society 10, 271-310.

NOYÉ 1994 = G. Noyé, Villes, economic et societe dans la province de Bruttium-Lucanie, nie du IV an VI siècle, in Francovich-Noyé 1994, 693-734.

NOYÉ 1998 G. Noyé, I centri del Bruzio dal IV al VI secolo, in Atti Taranto 1998, 431-470.

NOYE 2002 = G. Noye, Economia e società nella Calabria bizantina, in Placanica 2001, 577- 656.

OLEVANO- GUIDOBALDI 1993 = F. Olevano and F. Guidobaldi, Affermazione dei pavimenti in opus sectile in redazione marmorea, CCARB 40, 163-174.

OTRANTO 1991 = G. Otranto, Italia meridionale e Puglia paleocristiane. Saggi Storici, Ban.

OVADIAH 1980= A. Ovadiah, Geometric and Floral Patterns in Ancient Mosaics, Rome.

PACE 1951 = B. Pace, Villa di eta imperiale a Piazza Armerina, Gnomon 23, 469-471.

PACE 1955= B. Pace, I mosaici di Piazza Armerina, Roma.

306

PALOL 1989, P. “La arqueología cristiana en la Hispania romana y visigoda. Descubrimientos recientes y nuevos puntos de vista,” Actes du XI congres international d’archeologie chretienne,vol.2, Lyon.

PANELLA 1988.

PANELLA 1993 = C. Panella, Merci e scambi nel Mediterraneo tardoantico, in Storia di Roma, 3.2. I luoghi e le culture,” Torino 1993, 613-697

PANELLA – TCHERNIA. C. Panella, A. Tchernia, Produits agricoles transportés en amphores. L’huile et surtout le vin, in L’Italie d’Auguste à Dioclétien, Roma 1994, 145-165.

PAOLETTI 1994= M. Paoletti, Occupazione romana e storia delle città, in Storia della Calabria antica. Età Italica e Romana, S. Settis (ed.), Roma.

PARLASCA 1959 = K. Parlasca, Die römischen Mosaiken im Deutschland, Berlin.

PASSI PITCHER 1990a L.P. Passi Pitcher, Palazzo Pignano (Cremona). Loc. Pieve di S. Martino, BArch 1-2, 97-100.

PASSI PITCHER 1990b = L.P. Passi Pitcher, Il complesso di Palazzo Pignano: ii battistero, in Milano Capitale 1990, 266-67.

PERCIVAL 1976 = J. Percival, Tile Roman Villa. An Historical introduction, London.

PERGOLA 1999 = P. Pergola (ed), Alle origini della parrocchia rurale (1V-VIII sec.), Atti della giornata tematica dci Scm man di Archeologia Cristiana (Ecole Fran çaise de Rome, 19 marzo 1998), (Sussidi allo studio delle Antichità Cristiane 12), Rorna.

PERGOLA-SANTANGELI VALENZANI-VOLPE 2003 = P. Pergola-R. Santangeli Valenzani- R. Volpe (ed.), Suburbium. il suburbio di Roma dalla crisi del sistema delle ville a Gregorio Magno, (Collection de 1 ‘Ecole Française de Rome 311), Roma.

PIETRI 2002 = L. Pietri, Evergétisnie chrétien etjbndations privées dans l’Italie de l’Antiquite’ tardive, in Carrié-Lizzi Testa 2002, 253-263.

PISANI SARTORIO-CALZA 1976 = G. Pisani Sartorio-R. Calza, La villa di Massenzio sulla via Appia, Città di Castello.

PITTS 2006= M. Pitts, British Archaeology 91.

PLACANICA 1994= A.Placanica, Storia della Calabria, Roma.

307

PLACANICA 1999 = A. Placanica (ed.) La Calabria medievale, Culture, arti. tecniche, 2, Reggio Calabria.

REBILLARD 2003 = E. Rebillard, Religion et Sépulture: L'Église, les vivants et les morts dans l'Antiquité tardive.

REBILLARD 2009 = E. Rebillard, The care of the dead in Late Antiquity, Ithaca, Cornell University.

REBUFFAT 1970 = R. Rebuffat et alii, Thamusida, Fouilles du Service des Antiquités du Maroc, II, MEFR Suppl. 2, Rome.

Recueil III,1 = H. Lavagne, Recueil général des Mosaïques de la Gaule, III, Narbonnaise, 1, 1979.

Recueil III, 2 = J. Lancha, Recueil général des Mosaïques de la Gaule, III, Narbonnaise, 2, 1981.

Recueil IV,1= C. Balmelle, Recueil général des Mosaïques de la Gaule, IV, Aquitanie, 1, 1980.

RICH 1992 = J. Rich (ed.), The City in the Late Antiquity, London-New York.

Ripoll-Chavarria 2003 = G. Ripoll-A. ChavarrIa, Arquitectura religiosa en Hispania (siglos IV al VII). En forno a algunos nuevos hallazgos, Hortus Artium Medieva- hum 9, 95-112,

RIPOLL-ARCE 2000 = G. Ripoll-J. Arce, The Transfrrmation and End of Roman Villae in the West (Fourth-Seventh Centuries): Problems and Perspectives, in Brogiolo- Ziau1iier-Chjisfie’21Jt8D, 63-T14.

RODA 1985 = S. Roda, Fuga nel privato e nostalgia del potere nel IV secolo d. C.: nuovi accenti di un’antica ideologia, in Le trasformazioni della cultura nella Tardoantichità, Atti del Convegno ( 1982), Roma, vol. I, 95-110.

ROGERS 2006= C.F. Rogers, Baptism and Christian Archaeology, Oxford Clarendon (first edition: 1903).

ROLLER 2006 = M. Roller, Dining Posture in Ancient Rome: Bodies, Values and Status Princeton.

ROMIZZI 2006 = L. Romizzi, Le ville tardo antiche in Italia, in Chavarria-Arce-Brogiolo 2006, 37-60.

ROSEN-SLUITER 2006 = R. Rosen, I. Sluiter, City, Countryside, and the Spatial Organization of Value in Classical Antiquity, Leiden.

ROSSITER 1991 = J.J. Rossiter, Convivium and Villa in Late Antiquity, in W.J. Slater (ed.),

308

Dining in a Classical Context, Ann Arbor, 199-2 14.

ROSTOVZEV 1933 = M .1. Rostovzev, Storia economica e sociale del’impero romano, Firenze 1933 (first edition: Oxford 1926).

ROULAND 1979 = N. Rouland, Pouvoir politique et dépendance personelle dans l’antiquite romaine: genèse et role des rapport’s de clientele, (Collection Latomus 166), Bruxelles.

SABBIONE-BARELLO-BRIZZI-CARDOSA 1997 = C. Sabbione-F. Barello-M. Brizzi-M. Cardosa, I mosaici dell ‘edtficio termale in contrada Palazzi di Casignana (Reggio Calabria), in Bonacasa Carra-Gijidobaldi 1997, 385-400.

SALZMAN 2002 = M.R. Salzman, The Making of a Christian Aristocracy. Social and Religious Changes in the Western Roman Empire, Cambridge (MA) and London.

SALLER 1982 = R. Saller, Personal Patronage under the Early Roman Empire, Cambridge.

SANGINETO 1994 = A.B. Sangineto, Per la ricostruzione del paesaggio agrario della Calabria roinana, in Settis 1994, 558-593.

SANNAZARO 1992 = M. Sannazaro, Costituzione e sviluppo di centro religiosi cristiani nelle campagne lombarde: problemi topografici ed archeologici, in Brogiolo-Castelletti 1992, 61-71.

SARNOWSKI 1978 = T. Sarnowski, Les representations de villas sur les inosaIques africaines tardives, Wroclaw.

SCAGLIARINI CORLÀITA 1990 = D. Scagliarini Corlàita, Le grandi ville di età tardoantica, in Milano Capitale 1990, 257-262.

SCAGLIARINI CORLAITA et alii 1992= Villa Romana di Desenzano,Roma.

SCAGLIARINI CORLÀITA 1994 = D. Scagliarini Corlàita, La villa di Desenzano. Vicende architettoniche e decorative, in La Guardia 1994, 43-58.

SCAGLIARINI CORLÀITA 1995 = D. Scagliarini Corlàita, Gli ambienti poligonali nell’architettura residenziale tardoantica, in CCARB 1995, 837-873.

SCAGLIARINI CORLÀITA 1997 = D. Scagliarini Corlàita, La villa di Desenzano sul Garda, in Roffia 1997 a, 191-215.

SCAGLIARINI CORLÀITA 2003 = D. Scagliarini Corlàita, Domus, villae, palatia. Convergenze e divergenze nelle tipologie architettoniche, in Ortalli-Heinzelmann 2003, 153-172.

309

SCHEIDEL-VON REDEN 2002 = W. Scheidel and S.Von Reden (eds.), The ancient economy, New York.

SCHÖLLGEN 1995 = G. Schöllgen , “Balnea mixta: Entwicklung der Spätantiken Bademoral,” in Festschrift K., Thraede.

SCOTT 2004= S. Scott, Elites, Exhibitionism and the Society of the Late Roman Villa, in Christie 2004, 39-66.

SETTIS 1994 = S. Settis (ed), Storia della Calabria antica, II, Reggio Calabria.

SFAMENI 2005 = C. Sfameni, Le villae-praetonia: i casi di San Giovanni di Ruoti e di Quote San Francesco, in Volpe-Turchiano 2005, 609-622.

SFAMENI 2006 = C. Sfameni, Ville residenziali nell’Italia tardo antica, Bari.

SLATER 1991 = W.J. Slater (ed), Dining in a Classical Context, Ann Arbor.

SMALL 1980 = AM. Small, S.Giovanni di Ruoti: Some Problems in the Interpretation of the Structures, (British Museum Occasional Papers 24), London, 561-570.

SMALL 1981 = A.M. Small, The Environment of San Giovanni in the Roman Period, in G. Barker-R. Hodges (ed.), Archaeology and italian Society. Prehistoric, Roman and Medieval Studies, (Papers in Italian Archaeology, II, BAR mt. Sen. 102), Oxford, 203-212.

SMALL 1991 = A.M. Small, Late Roman Rural Settlement in Basilicata and Western Apulia, in Roman Landscapes. Archaeological Survey in the Mediterranean Region, London, 204-222.

SMALL 1998 = AM. Small, La Basilicata nell’etâ tardoantica: ricerche archeologiche nella valle del Basentello e a S. Giovanni di Ruoti, in Atti Taranto 1998, 33 1-342.

SMALL 2005 = A. Small, Le analisi al radiocarbonio e la fine della villa di S. Giovanni di Ruoti, in Volpe-Turchiano 2005, 127-132.

SMALL-BUCK 1980 = A.M. Small-R.J. Buck, The Topography of Roman Villas in Basili- cata, in Attività archeologica in Basilicata 1964-1977, Scritti in onore di D. Ada- mesteanu, Matera, 561-570.

SMALL-BUCK 1994 = AM. Small-R.J. Buck, The Excavations of San Giovanni di Ruoti, I. The Villas and their Environment, (, Suppl. volume 33), Toronto.

SMITH 1997 = J.T. Smith, Roman Villas. A Study in Social Structure, London and New York.

310

SODINI 1995 = J.P. Sodini, Habitat de l’Antiquité tardive, Topoi 5, 151-218.

SODINI 1997 = J.P. Sodini, Habitat de l’Antiquité tardive, 2, Topoi 7, 2,435-577.

SOGNO 2006 = C. Sogno, Q. Aurelius Symmachus: A Political Biography, 2006.

SPADEA 1989 = R. Spadea (ed.), Da Skvlletion a Scolaciuni. Ilparco archeálogico del.P. la Roccelletra, Roma.

SPERA 1999 = L. Spera, Il paesaggio suburbano di Roma dall’antichità al Medioevo. Il comprensorio tra le vie Latina e Ardeatina dalle Mura Aureliane al 111 miglio, Roma.

SPIDLIK 1996 = T. Spidlik, Melania. La benefattrice (383-440), Milano.

Suburbium 2000 = Suburbium. Dalla crisi del sistema delle ville a Gregonio Magno, Giornate di studio sul suburbio romano (Roma 16-18 marzo 2000), Roma.

SWOBODA 1919 = K. M. Swoboda, Romische und Romanische Paläste, Wien.

TAMM 1963 = B. Tamm, Auditorium and Palatium, Stockholm.

THEBERT 1986 = Y. Thebert, Vita privata e architettura domestica nell’Africa romana, in P. Ariès-G. Duby (ed.), La vita privata dall’impero romano all’anno mille, Roma Bari 1986, 233- 309.

TODD 1978 = M. Todd (ed.), Studies in the Romano-British Villa, Leicester.

TODD 2005= M. Todd “Baths or Baptisteries? Holcombe, Lufton and Their Analogues,” Oxford Journal of Archaeology 24 (3) pp. 307-311.

VAQUERIZO-CARRILLO DIAZ-PINES 1995 = The Roman Villa of El Ruedo (Almedinilla, Córdoba), JRA 8, 121-154.

VERA 1983 D. Vera, Temi e problemi della villa di Piazza Armerina, in AA.VV. 1983, 581-593.

VERA 1986 = D. Vera, Simmaco e le sue proprietà: strutture e funzionamento di un patrimonio aristocratico del quarto sec. d.c., in Paschoud 1986, 231- 270.

VERA 1987 = D. Vera, Colonato e trasformazioni agrarie nell’Africa proconsolare nel tardo impero, in A. Mastino (ed.), L’Africa romana, Atti del IV Convegno di studio (Sassari 12-14 dicembre 1986), Sassari, 267-293.

VERA 1988 = D. Vera, Aristocrazia romana ed economie provinciali nell’Italia tardoantica: il caso siciliano, QC X, 19, 115-172.

311

VERA 1994a = D. Vera, Conductores domus nostrae, conductores privatorum. Concentrazione fondiaria e redistribuzione della ricchezza nell’Africa tardoantica, in Istitutiones, societe et vie politique dans l’Empire Romaine au IV siècle ap. J.C., Actes de Ia table ronde autour de l’ouvre d’AndrC Chastagnol (Paris 20-21 janvier 1989), Roma, 465-490.

VERA 1994B = D. Vera, L’Italia agraria nell’età imperiale fra crisi e trasformazione, in L’Italie d’Auguste a Diocletian 1994, 239-248.

VERA 1994C = D. Vera, Il sistema agrario tardoantico: un modello, in Francovich-Noyè 1994, 136-138.

VERA 1999a = D. Vera, I silenzi di Palladio e l’Italia, AntTard 7, 283-297.

VERA 1999b = D. Vera, Massa fundorum. Forme della grande proprietà e poteri della cittâ in italia Ira Costantino e Gregorlo Magno, MEFRA 111, 2, 991-1025.

VERA 2001 = D. Vera, Sulla riorganizzazione agraria dell’ Italia meridionale in età imperiale: origini, forme e funzioni della massa fundorum, in Lo Cascio-Storchi Marino 2001, 613-634.

VERA 2005= D. Vera, I paesaggi rurali del Meridione tardo antico: bilancio consuntivo e preventivo di Domenico Vera, 23-38, Bari.

VEYNE 198(ed.), A History of Private Life: I. From Pagan Rome to Byzantium , 1987-1991, pp. 371-374;

VITTI 1993 = M. Vitti, Il Palazzo di Galerio a Salonicco, RTopAnt 3, 77-106.

VOLPE 1996 = G. Volpe, Contadini, pastori e mercanti nell’Apulia tardoantica, (Munera 6), Bari.

VOLPE 1998 = G. Volpe, Paesaggi della Puglia tardoantica, in Atti Taranto 1998, 267-330.

VOLPE 2000 = R. Volpe, Le ville del suburbio di Roma, in Ensoli-la Rocca 2000, 161-167.

VOLPE-DE FELICE-TURCHIANO 2004 = G. Volpe-G. De Felice-M. Turchiano, Musiva e sectilia in una lussuosa residenza rurale dell ‘Apulia tardoantica: la villa di Faragola(Ascoli Satriano), Musiva&Sectilia, 127-158.

VOLPE-DE FELICE-TURCHIANO 2005a = G. Volpe-G. De Felice-M. Turchiano, I rivestimenti rnarmorei, i mosaici e i pannelli in opus sectile vitreo della villa tardoantica di Faragola (Ascoli Satriano Foggia), in Atti del X Colloquio AISCOM ( 18-21 febbraio 2004), Roma, 61-78.

312

VOLPE-DE FELICE-TURCHIANO 2005b = G. Volpe-G. De Felice-M.-Turchiano, Faragola (Ascoli Satriano). Una residenza aristocratica tardoantica e un ‘villaggio’ altomedievale nella valle del Carapelle: primi dati, in Volpe-Turchiano 2005, 265-298.

VOLPE-TURCHIANO 2005 = G. Volpe-M. Turchiano (ed.), Paesaggi e insediamenti rurali in Italia meridionale fra Tardoantico e Altomedioevo, Atti del Primo Seminario sul Tardoantico e l’Altomedioevo in Italia meridionale, (Foggia, 12-14 febbraio 2004) (Insulae Diomedeae 4), Bari.

VON BOESELAGER 1983 = D.V. Boeselager Antike Mosaiken in Sizilien, Rome.

VON STACKELBERG 2009= K. T. VON STACKELBERG, The Roman garden : space, sense, and society, New York

VOZA 1976-77 = G. Voza, L’attività della Soprintendenza alle Antichità della Sicilia Orientale, Kokalos 22-23, 2, 572-579.

VOZA 1980-81 = G. Voza, L’attività della Soprintendenza alle Antichità della Sicilia Orientale, Kokalos 26-27, 2.1, 690-693.

VOZA 1982 = G. Voza, Attività della Soprintendenza ai Beni Archeologici della Sicilia Orientale dal 1976 al 1982, BCASic 3,93-137 (111-121).

VOZA 1984-85 = G. Voza, Attività nd territorio della Soprintendenza a/ic Antichità di Siracusa nd quadriennio 1980-84, Kokalos 30-31, 659-661.

VOZA 1989 = G. Voza, I crolli della villa romana di Patti Marina, in E. Guidoboni (ed), I terremoti prima del 1000 in Italia e nell’area mediterranea,Bologna, 496-501.

VOZA 1999 = G. Voza, Nel segno dell’antico. Archeologia nel territorio di Siracusa, Palermo, 121-129.

VOZA 2003 = G. Voza (ed.), I mosaici del Tellaro. Lusso e cultura nd sud-est della Sicilia, Siracusa.

WALLACE-HADRILL 1988 = A. Wallace-Hadrill, The Social Structure of the Roman House,PBSR 56, 43-97.

WALLACE-HADRILL 1998 = A. Wallace-Hadrill, The Villa as Cultural Symbol, in Frazer 1998, 43-53.

WARD-PERKINS 1984 = J.B. Ward-Perkins, Etruscan and Roman Architecture, Harmondsworth.

WARD-PERKINS 1974= J.B. Ward-Perkins, Architettura Romana, Milano.

313

WARD-PERKINS 1996 = B. Ward-Perkins, Urban Continuity?, in N. Christie-S .T. Loseby (ed.), Towns in Transition. Urban evolution in Late Antiquity and the Early Middle Ages, Aldershot, 4-17.

WEBER 1996 = M. Weber, Antike Badekultur, Munich.

WESTGATE 2000= R. Westgate, “Pavimenta Atque Emblemata Vermiculata: Regional Styles in Hellenistic Mosaic and the First Mosaics At Pompeii,” American Journal of Archaeology 104 (2), 2000, pp. 255–275.

WHITTAKER, C.R., “Rural life in the later Roman empire” in The Cambridge Ancient History Vol. 13: The Late Empire, A.D. 337–425.

WICKHAM 1994 = C. Wickham, Land and Power. Studies in Italian and European Social History, 400-1200, London.

WICKHAM 2002 C. Wickham, Rural Economy and Society, in La Rocca 2002, 118-143.

WICKHAM, C., Framing the early Middle Ages: Europe and the Mediterranean 400-800, Oxford 2005

WILKES 1986 = J.J. Wilkes, Diocletian’c Palace, Split: Residence of a Retired Roman Emperor, Sheffield.

WILSON 1983a = R.J.A. Wilson, Piazza Armerina, London.

WILSON 1983b = R.J.A. Wilson, Luxury Retreat, Fourth-Century Style: a Millionaire Aristocrat in Late Roman Sicily, in AA.VV. 1983, 537-552.

WILSON 1984 = RJ.A. Wilson, Piazza Armerina and the Senatorial Aristocracy in Late Roman Sicily, in Rizza-Garraffo 1984, 170-1 82.

WILSON 1990 = R.J.A. Wilson, Sicily under the Roman Empire, The Archaeology of a Roman Prooince 36 BC-AD 535, Warminster.

YASIN 2005 = A.M. Yasin, Funerary Monuments and Collective Identity: From Roman Family to Christian Community, The Art Bulletin, 87, no. 3.

YEGÜL 1992= F. Yegül, Bath and Bathing in Classical Antiquity, London.

YEGÜL 1999= F. Yegül, “Bathing,” in G.W. BOWERSOCK, P.R.L. BROWN, O. GRABAR (ed.), Late Antiquity. A guide to the Postclassical World, London

ZAHN 1991 = E. Zahn, Die Basilica in Trier: Romisches Palatium. Kirche zum Erlöser, Trier.

314

ZANKER 2002 = P. Zanker, Domitian’s palace on the Palatine and the imperial image,” in A. K. Bowman et alii (ed.), Representations of the Empire. Rome and the Mediterranean World (Oxford 2002), pp.107-130.

ZINZI 1998 = E. Zinzi, Siti e memorie cassiodoree in Calabria, in Atti Taranto 1998, 37-65.

315